Psychological Testing

PSYCHOLOGICALTESTING
Presented by:
ROBERT GORDON, ID., PH.D., Addison

Wilmington Institute Network
Writtenby:
MARY JOHANNA MCCURLEY, Dallas
M'Curley, Orsinger, M'Curley, Nelson & Downing, L.L.P.
I
KATHRYN MURPHY, Plano
Koons, Fuller, Vanden Eykel & Robertson
*Reprinted from: AmericanAcademyof Matrimonial Lawyers
March, 2005
State Bar of Texas

31ST ANNUAL ADVANCED FAMILY LAW COURSE
August 8-11, 2005
Dallas
CHAPTER 5
WIN
WILMINGTON INSTITUTE NETWORK
Trial and Settlement Sciences
DOCTOR ROBERT GORDON
Dr. Robert Gordon is a forensic and clinical psychologist and

attorney. He is founder and director of the Wilmington
Institute of Trial and Settlement Sciences. He is the pioneer
in the field of trial and settlement psychology in America. His
study of the American Jury Process began in 1968. He has
provided consultation on many of the celebratedAmericanjury
trials during the past 20 years. Dr. Gordon's vision of this field
for the 2l Century is to contribute to the administration of
justice through psychological research conducted in a powerful
and ethical manner.
Dr. Gordon is boarded in the specialties of clinical and forensic

psychology by the American Board of Professional Psychology.
He earned a Ph.D. degree in psychology from the University of
Oklahomaand a J.b. degree in law from Baylor University. Dr.
Gordon has served as Chair of the Texas State Board of
Examiners of Psychologists. He is the past president of the
Texas Psychological Association. In 1991, his work in
establishing this new field was recognized by the Texas
Psychological Foundation in the granting of the Distinguished
PsychologistAward.
The Wilmington Institute provides trial and settlement science

services in State and Federal forums, as well as assisting
international clients with domestic cases, The Institute's
facilities provide hi-tech, ADR and courtroom environments
with innovative information systems in Dallas and Houston,
Texas.
The Institute has conducted extensive research in order to
better understand how and why jurors make decisions in
complex civil lawsuits. His work has contributed to the
development of dispute resolution procedures such as summary
jury trials and other innovative methods for settling important
cases.
The work of Dr. Gordon and the Institute Team has been
featured on the MacNeil-Lehrer News Hour, ABC Nightline,
CNN News, Larry King Live, Good Morning America, U.S.A.
Today, Sky-T.V., Reuters News and in leading Japanese
Newsprint. He has frequently been a guest commentator on
legal issues for CNN, Fox News, and MSNBC.
Dr. Gordon lives in both Dallas and HoustonTexas with his wife Susan
who is a teacher. The Gordon's have two children. Ami Gordon who is a
practicing attorney in San Antonio and married to Scott Henslee who
is a dermatologist. The Gordon'sson Adam is a teacher who is on a
year's sabbatical teaching in Japan.
Wilmington Institute & Lw. Robert Gordon 2

MARYJOHANNA MCCURLEY
M"Curley, Orsinger,McCurley, Nelson & Downing, L.L.P.
5950 Sherry Lane, Suite 800
Dallas, Texas 75225
Phone: 214/273-2400
marvio(E!momnd.com
MARYJOHANNAMCCURLEY: born Baton Rouge, Louisiana, October3, 1953; admitted

to bar, 1979, Texas.
EDUCATION:
Centenary College (B.A., 1975).
Louisiana State University; St. Mary's University of San Antonio (J.D., 1979).
Board Certified, Family Law, Texas Board of Legal Specialization (1984).
EXTRACURRICULAR RESPONSIBILITIES:
Delta Theta Phi (Vice President, 1978-1979).
President, Student Senate, St. Mary's School of Law, 1979.
PROFESSIONAL RESPONSIBILITIES:
Assistant Editor, Family Law Manual, State Bar of Texas, 1982-1989.
Co-Editor Family Law Manual, State Bar of Texas 1989-1990.
Course Director, American Academy of Matrimonial Lawyers, MatrimonialLaw - 1998,
1998.
FACULTY:
National Business Institute, Inc. - Child Custody and Child Visitationin Texas, 1992.
Co-Chairperson, Texas Supreme Court Child Support and Child Visitation Guidelines
Committee, 1991 and 1992.
CourseDirector, Life, Lawyeringand the Pursuit of Happiness, State Bar of Texas, 1995.
Member, Texas SupremeCourt Child Supportand Child Visitation GuidelinesCommittee,
1996-1997.
CourseDirector, Art and Advocacyof Family Law, State Bar of Texas, 1997.
Assistant CourseDirector, Advanced Family Law Course, State Bar of Texas, 2000.
MEMBER:
The Best Lawyers in America - 2005-2006edition
State Barof Texas - Chair of Family LawSection of the State Bar of Texas, 2003 (Adjunct
Member, Family Law Council, 1982-1985: Family Law Council, 1987 - present;
Dallas (Family Law Section, Board of Directors, 1981-1982; Secretary-Treasurer,
1983; Vice Chairman, 1984; Chairman 1985; Past Chairman, 1986).
American Academyof MatrimonialLawyers (AAML - Texas Chapter - Treasurer, 1993-
1995; Secretary, 1995-1996; President-Elect, 1996-1997; President, 1997-1998)
(AAML - National - Secretary, November 2000 - 2002, Vice President 2002 -
present.)
Annette StewartAmerican Inn of Court- Masterand Secretary/Treasurer, 2003-2004
Special Concernsfor Children Committee, 1997-present and Co-Chair of this Committee,
2001 and 2002.
Co-Chairperson, Continuing Legal Education, 1995 and 1997-1998.
Member,Interdisciplinary Relations - Mental Health, 1997- present.
Member,Board of Governors, 1998 - present.
Member, Advertising Committee; Member, Marketing and Public Relations Committee.
Member, Stepfamilies Rights and Adoption Committee, 1994.
Chairperson, Mediation Study Committee, 1993.
American Bar Association.
Chairman, AssociateJudge'sTraining, 1998, 1999.
Board of Directors, Texas Academy of Family Law Specialists, 1990-1992, current
member.
AUTHOR AND LECTURER:

"Psychological Testing" co-authored with Jonathan Gould, Ph.D. and Kathryn Murphy,
AmericanAcademyof Matrimonial Lawyers Annual Meeting, March 5, 2005.
"Relocation Debate" co-authored with Kathryn Murphy, Advanced Family Law Course -
State Bar of Texas, August, 2004.
"Two People, Two Professions, One Marriage" co-authored with Mike McCurley, Fort
Worth Business Press, June of 2004
"Family Law: Changes to Keep in Mind" co-authored with Laura M. Hilliard, Texas Bar
Journal,January2004 edition
"Did Your Client 'Get Hitched'WithoutA Hitch - Defendingand Attacking Premarital and
Post-Marital Agreements" co-authored with Carson P. Epes, NewFrontiers in Marital
Property Law - State Bar of Texas, 2003
"Don't Forgetthe Child" - Lecture at Advanced Family Law Course- State Barof Texas,
2003
"Top Ten Things You Should Know to Ask About Handling Kid Cases" co-authored with
Laura M. HiUiard, Marriage Dissolution Course- State Bar of Texas, 2003
"Love and Money - Why More People are Seeking Prenuptial Agreements in the
Millennium" coauthored with CarsonP. Epes, Fort Worth Business Press, October
3, 2003
"Instant Message, Instant Email, Instantly Gone? - The Roleof Technology in Infidelityand
Divorce" co-authored with Laura M. Hilliard, Fort Worth Business Journal,October
3, 2003
"Maintenance is Alive and Well in Texas" co-authoredwith A. Michelle May, Advanced
Family Law Course- State Bar ofTexas, 2002
"A Primeron Psychological Testing,"Advanced Family Law Course- State Bar of Texas,
2001.
"Reimbursement Issues" co-authoredwith Drew Ten Eyck, MarriageDissolution Course-
State Bar of Texas, 2000.
"Shrinking the Shrinks Down to Size" co-authoredwith Mike McCurley and DrewTen Eyck,
AdvancedFamily Law Course- State Bar of Texas, 1999.
"Defending Against Malpractice" co-authored with JuliePruett,MarriageDissolution Course
- State Bar of Texas, 1999.
"Shrinking theShrinks Down to Size"co-authoredwith Mike McCurley, ChrisLake and Ken
Rockenbach, Advanced Family Law Course - State Bar of Texas, 1998.
"Effectively Deafing with The Child's Choice of Managing Conservator" co-authoredwith
R. Scott Downing and Kenneth W. Rockenbach, Texas Academyof Family Law
Specialists - State Bar of Texas, 1998.
"Effectively Dealing with the Child's Choice of Managing Conservator," co-authoredwith
R. Scott Downing and Kenneth W. Rockenbach, Texas Academyof Family Law
Specialists - State Bar of Texas, 1997.
"Preparing for the Petitioner's Financial Temporary Hearing," co-authoredwith R. Scott
Downing, Marriage Dissolution Course- State Barof Texas, 1997.
"Dealing with Ad Utems," co-authoredwith Jack W. Marr & Kathleen Cardone, Advanced
-
Family LawCourse State Bar of Texas, 1997.
"Pensions, StockOptions and Other NewSourcesofWealth," AnnualDivorce Conference
of the Dallas Chapter, Texas Society of Certified Public Accountants, 1997.
"Attorney Ad Litem and Guardian Ad Litem Practice," (co-authorwith ReginaldA. Hirsch
and R. Scott Downing), Advanced Family Law Course- State Bar of Texas, 1996.
"Gender Bias in Our Courts and Practice - Fact or Fiction?," Advanced Family Law
Course - State Bar of Texas, 1995.
"Peculiar Characterization Issues," 18th Annual Marriage Dissolution Course- State Bar
of Texas, 1995.
"An Overview of Psychological Testing," Illinois State Bar Association Family Law
Handbook, 1994.
"An Overview of Psychological Testing," Advanced Family Law Course - State Bar of
Texas, 1994.
"Psychological Testing and the ExpertWitness,"The Practical Lawyer, October 1994.
"An Overview of Psychological Testing," Utah chapter of the American Academy of
Matrimonial Lawyers, 1994.
"An Overview of Psychological Testing," American Bar Association Family Law
Conference, 1993.
"Discovery," Marriage Dissolution Course- State Bar of Texas, 1993.
"Social Studies and Psychological Evaluations: Their Use in Evidence and Howto Cross
Examine," Advanced Family Law Course- State Barof Texas, 1993.
'We Have -
Found The Enemy It Is Ust" or Self Management, Not Stress Management,
AdvancedFamily Law Course- State Bar of Texas, 1993.
"Tracing," State Bar of Texas, Advanced Family Law Course- State Bar of Texas, 1992.
"Child Support Issues - Modification, Second Family and Other Significant Problems,"
South Texas College of Law, 1992.
-
"Ethics," State Bar ofTexas, AdvancedFamily Law Course State Bar of Texas, 1992.
"Modification, From Sole to Joint and Back Again," Annual Family Law Institute, South
Texas College of Law, 1990, 1992.
"Managing Stress," New York BarAssociation, 1991.
"Attorneyand Guardian Ad Litems,"Advanced Family Law Course- State Bar of Texas,
1991.
"DivorceTaxation," American Academy of Matrimonial Lawyers, 1991.
"Dealingwith Experts and Psychological Tests," Marriage Dissolution Course- State Bar
of Texas, 1991.
"Alimony,' co-authored with JimmyL. Verner, Jr., Advanced Family Law DraftingCourse-
State Bar ofTexas, 1990.
'innovative Ideas for Setting and Collecting Attorney Fees," American Academy of
Matrimonial Lawyers, 1990.
-
"Managing Stress,"Advanced Family Law Course State Bar ofTexas, 1990.
"Psychological Testing: An Update," Marriage Dissolution Course, State Bar of Texas,
1989.
"Enforcement- Support and Visitation", Marriage DissolutionCourse- State BarofTexas,
1986.
"Jury Selection in Custody Cases," Advanced Family Law Course- State Bar of Texas,
1987.
"Evidentiary Issues in Child Support, Texas Revolution in Child Support: Perspective,
Pleading and Practice,"State Bar of Texas, 1986.
KATHRYNJ. MURPIIY
KOONS,FULLER,VANDENEYKEL& ROBERTSON,P.C.
5700 W. Piano Pkwy., Suite 2200
Piano, Texas 75093
(972) 769-2727
Internet: Kathrvn(ñ)koonsli;ller.corn
Website: www.koonsliiller.com
EDUCATION:
J.D. 1989 SouthernMethodist University,Dallas, Texas
BA 1986 University ofTexas at Tyler, Tyler, Texas
PROFESSIONAL ACTIVITIESANDAFFILIATIONS:
Partner - Koons, Fuller, Vanden Eykel, & Robertson P.C.
Board Certified Family Law - Texas Board of Legal Specialization
SPECIAL RECOGNITIONS/HONORS:
Listed in D Magazine "BestLawyers in Dallas" for Family Law (May 2001 and May 2005)
Listed in "ilie Best Lawyers in America"
Listed in TexasMonthly— Texas Super Lawyers (Top 50 Women Attorneys in Texas; Top 100 Attorneys in Texas)
(2003)
Listed in Texas Monthly— Texas Super Lawyers (Top 50 Women Attorneys in Texas; Top 100 Attorneys in Texas)
(2004)
Fellow - American Academy ofMatrimonial Lawyers

Member - Texas Academy of Family Law Specialists
Member - Family Law Council,StateBar ofTexas
Member - Dallas AllianceofCollaborativeFamily Lawyers
Member - Collaborative Law Institute ofTexas
Member - International Academy ofCollaborative Professionals
Member - State Bar ofTexas(Family Law Section)
Member - American Bar Association(Family Law Section)
Member - College ofthe State Bar ofTexas
Member - Dallas Bar Association (Family Law Section)
Member - Collin CountyBar Association(FamilyLaw Section)
Member - PIano Bar Association
Co-Chair - DAYL People's Law School (1995 & 1996)
Member - J. Reuben Clark Law Society - Board of Directors (1997-2000)
Instructor - Southeastern Paralegal Institute (1993-1996)
Member - Fee Dispute Committee - Dallas Bar Association (1998 -2000)
Fellow - Collin CountyBenchBar Foundation
Chairman - Program Committee,Collin County Bench Bar Conference. 2000
Board Member- AdvisoryBoard in CASA
Co-Chair - 2001 Collin County Bench Bar Conference
Member - Grievance Committee, District IA (2002 - present)
Certificate - Advanced Mediation - Family Law (1995)
Partner - Dallas Social Venture Partners (2003 - present)
PUBLICATIONS:
Co-Author, West Publishing Company, TEXASFAMILY LAW PRACTICE GUIDE(published March 2000,
supplemented 2001, 2002, 2003 and 2004)
Author, PROTECTING YOURASSETS FROMA TEXASDIVORCE(2005).

PtJBLICATIC)NS/LECTURES:
CHILD SUPPORT, 32" Annual Advanced Family Law Course, State Bar ofTexas(August 2005)
PSYCHOLOGICAL TESTING, American Academy of Matrimonial Lawyers (March 2005)
RELOCATION, Dallas VolunteerAttorney Program and Family Law Section, 'Nuts and Bolts" Family Law
Training (February 2005)
OPENING STATEMENTSANDCLOSING ARGUMENTS, Ultimate Trial Notebook— Family Law, State Bar of
Texas(December 2004)
HIGH TECHEVIDENCE, WEBSITES, HAItDDRIVES, if-MAILS, E, Williamson County (October2004)
RELOCATION DEBATE, 30"' Annual Advanced Family Law Course, State Bar ofTexas (August 2004)
COLLABORATIVELAWPANEL, Collaborative Law Spring Retreat (March 2004)
HIGH TECH EVIDENCE, WEBS!TES, HARDDRJyES, E-MAILS, ETC, 29th Annual Advanced Family Law
Course, StateBar ofTexas (August 2003)
FAMILYLAWEVIDENCE, Collin County Bench Bar Conference (May 2003)
FAMJLYLAJVEVIDENTIARYISSUES, 16°' Annual Advanced Evidence & Discovery Course, State Bar of
Texas (March 2003)
PARENTAGE: ESTABLISHING, ATTACKING& UNDOING, Texas Academy ofFamily Law Specialists
(January2003)
PARENTAGE: CURRENT ISSUES, University of Houston Law Cenler (October 2002)
OBJECTIONS AT TRIAL,American Bar Association, Section ofFamily Law, Fall CLEConference, Orlando, FL
(November 2002)
DIVISION OF PROPERTYAND DIVIDING SPECIFICASSETS, University of Houston Law Foundation (June
2002)
FAMILYLAWFORTHE NON-SPECIALIST,J. Reuben Clark Law Society (April 2002)
DIVISIONOF PROPERTYON DIVORC'E, University of Houston Law Foundation (September 2001 and
September 2002)
FAMILYLAW UPDATE,Ten Hot Topics in Family Law, Collin County Bench/Bar Conference (June 2000)
MODERATOR, Collin County Bench/BarConference (June 2000)
FAMILYLAW UPDATE,J, Reuben Clark Law Society (June2000)
DIVISION OF PROPERTYON DIVORCE,University ofHouston Law Foundation (April 2000)

FAMILYLAJVDISCUSSION,J. Reuben Clark Law Society(October, 1999)
CROSS-EXAMINAHON OFA VALUATIONEXPERTAND BUSINESS VALUATIONSINDIVORCE,
American Bar Association,Family Law Section, San Diego (October, 1999)
OPENINGSTATEMENT-JURYOR NONJURY, 25" Annual Advanced Family Law Course, State Bar of
Texas (August, 1999)
UPDATE ONFAMILYLAW, Dallas Women'sLawyers Association (December, 1998)
FAMILYLAWFORTHE NON-SPECIAL1ST, Dallas Bar Association (December, 1998)
OBTAININGRECORDS FROMNON-PARTIES, 24th Annual Advanced Family Law Course, State Bar of
Texas (August 1998)
DISCOVERY UPDATE,Dallas Bar Association Friday Clinic (December 1996 and July 1998)
OPENING ANDCLOSING THE FILE, Family Law Conference forthe General Practitioner and Legal Assistant,
South TexasCollege ofLaw (February 1998 and February 1999)
DISCOVERY UPDATE,23rd Annual Advanced Family Law Course, State Bar ofTexas (August 1997)
DISCOVERY UPDATE,22ndAnnual Advanced Family Law Course, State Bar ofTexas(August 1996)
BUSINESS VALUATIONINDIVORCE, Dallas Chapter Texas Society ofCertified Public Accountants Fifth
Annual Divorce Conference (September 1996, September 1998)
DISCOVERY, Dallas Bar Association Family Law Training Seminar (October 1996)
ATTORNEY'S FEES, Family Law Practice Institute, University ofHouston (September 1996)
PsychologicalTesting Chapter 5
TABLE OF CONTENTS
I. INTRODUCTION I
U. GENERAL CONSIDERATIONS OF PSYCHOLOGICAL TESTING

A. Reliability
B. Validity
Ill. INTELLIGENCE TESTS 2
A. Purpose ofIntelligence Tests 2
B. Stanford-BinetIntelligenceScale 3
C. WechslerAdult Intelligence Scale- Revisedor Third Edition(WAIS-R) 4
IV. PERSONALITY TESTS 8

A. ObjectivePersonality Tests 8
B. I6PF 18
V. NEW DEVELOPMENTS IN PSYCHOLOGICAL TESTING WITHtNTHE LAST DECAl) 29

A. Bricklin PerceptualScales (Bps) 29
B. Sex Abuse Legitimacy Scales(Sals) 33
VI. CONCLUSION 54
A. GeneralConsiderationsWhen an Experthas Psychological Testing 54
B. Standards and Guidelinesfor Child CustodyEvaluations in Divorce Proceedings 55
C. Final Thoughts 55
Psychological Testing Chapter 5
PSYCHOLOGICALTESTING
I. INTRODUCTION
The family law practitioneris confronted with the resultsof psychological testing with increasing frequency.
Thetests aregenerallyinterpreted as establishingthe mental healthand parenting skills ofa parent.
Psychological testingbecamepopularduringWorld War11, and clinicalpsychologistshavebeen toting testkits
eversince. Asthis article illustrates, psychologicaltests havea potentiallyvaluable role to play intheassessmentofthe
mental health ofparentsand ofchildren. However,more problematic is the applicationofpsychologicaltestingto the
resolution of custody issues. Because all psychological tests are eventually filtered through the psychologist's
interpretation,the tests can be no better or worse than the interpretation itself.
This article supports the continued development, administration and use ofpsychologicaltesting in custody
disputes, but it also sends a warning to attorneys to not always have faith that these tests are valid, fair, or even
constitutionallyappropriate. As David Viscount, in his book, The Making ofa Psychiatrist, observes, "Mental health
professionalslovetheirinterpretations, they are gifts that they give themselves.'
II. GENERAL CONSIDERATIONSOF PSYCHOLOGICAL 1'ESTING
Psychological testingis a standardized methodofcheckingaportionofan individual'sbehavior and comparing

it to that ofa groupwith knowncharacteristics. Thetests arecategorizeddepending on what factorsthe particulartest
is designedto measure. Intelligence tests,personalitytests, achievement tests,andneuropsychologicaltests haveallbeen
developed in an effort to create objective and standardizedinstruments. [D. Shuman,Psychiatricand Psychological
Evidence47(1986).] Since cases involvingthe parent-childrelationship involve psychologicalissues and, therefore,
often requirepsychologicaltestimony regardingintelligenceand personality,this article will examinethe major tests in
these areas.
In appraisinga particularpsychologicaltest, there are two considerations whichmust always be keptin mind:
reliabilityand validiti' ofthe particulartest.
A. Reliability
Reliabilityofa given test requires consistencyofresults. For example,in order for a lest to be "reliable" in
scientific terms, the test results should be essentiallythe same whether the patient is tested six days, six weeks or six
months later.
The legal analogyofreliability is witness credibility. IfMr. Jones saysin adepositionthat Mrs. Smith is a good
mother, but at trial he says she is not always a good mother, then he is an "unreliable" witness. What was true at the
depositionshould alsobe trueattrial, Ifsomeoneis tested on August 1,2004 andagain on January 1,2005, thenthe test
resultsshould be essentiallythe same ifthe test is reliable.
Reliabilityofa testisusuallydetermined in termsofa coefficient correlation betsveen test scoresona firsttesting

and then on a second. A reliabilitycoefficientvalue ranges from 0.0 (no reliability)to +1.0 (perfectreliability). There
is not complete agreement among psychologistsas to what reliabilitystandardshouldbe set in order to be confidentof
a certaintest. "Correlation coefficient" means, in statisticalterms,the ratiobetweentwo differentquantitativemeasures,
in this situationrepresentingthe degree ofassociationofthevariablephenomenaofa first testtaking versus a second.
In laymen'sterms,this meansthat the closerthe resultsare between the firsttestingand the second testing, the closerto
thecoefficientofa 1.0 theresultswill be. The more conservativegroup believes a correlationcoefficientof0.90is the
lowestacceptablecorrelation coefficientforatestto be considered "reliable." Anotherschoolofthought arguesthat tests
with areliabilitycoefficientofo.80aresufficient.Clearly,a testwith less than a 0.80reliabilitycoefficientdoesnotmeet
thestandardsofreliabilityaccepted by eitherschoolofthought. [J. Ziskin,Copingwith Psychiatricand Psychological
Testimony,Vol. 1, 211(1981).] However,the attorney should be aware that for every psychologistand psychiatrist
existing, there is a differentopinion as to what the lower limits ofreliabilityoughtto be.
Il. Validity
Validationofa test, for scientificpurposes,requires that the test measure what it actuallypurports to measure.
Forexample,does a particularintelligencetesttruly measureintelligence? TheAmericanPsychologicalAssociation,
along with two other associations,haveproduceda report entitled,"Educationaland Psychological Tests and Manuals,
j." The report is designedto help the clinical psychologistdetermine"validity" of a certaintest.
PsychologicalTesting Chapters
However,validity is more difficult to establish thanreliability. For example,assumefive witnessesalt testify

that Mr. Jones is a good businessman (highreliability). Is thererelevanceto that undisputedtestimony,when the issue
is whether he is cruel to his wife? In other words, no valid relationship has been establishedbetween being a good
businessman and mentalcruelty.
There are differentapproachesto validity, whichare:

I. "PREDICTIVE VALIDITY - The most commonly used and useful validation process is
predictivevalidity. This means that with certain information about A, one can state with
probabilitythat B willoccuriljat 78.)
2. "CONCURRENT" VALIDITY- Concurrent validity is essentially the same as predictive
validity, except insteadofbeing predictive ofbehavior,the test scoresare compared to present
3.
(
information. Comparingwhat scoreonemakesonan intelligencetest with what gradesthe test
taker is making in school is an example ofconcurrentvalidity. at 80.)
"CONTENT" OR "FACE"VALIDITY -Thiskind ofvalidity is basedon what we call in law,
res ipsa loguitur, "the thing speaksfor itself." If the test involvedproblemsin multiplication,
it would then seem to follow that the test measuresone's ability to multiply.
4. "CONSTRUCT" VALIDITY- This is an all-encompassingconceptwhich assimilatesone or
more of the other types of validity. "It refers to the extent to which the test may be said to
measure a theoretical concept, for example, intelligence or mechanical comprehension or
anxiety. Itentails abroader,more enduringand more abstractkind ofbehavioraldescription."
(Id. at 81.)
ilL INTELLIGENCE TESTS

A. Purposeof Intelli2enceTests.
Intelligence tests are astandardizedmethod ofevaluating mentalability,whichaids the clinicalpsychologistin
thediagnosis of individuals. Intelligence tests are used to achieve four main purposes:
I. An Appraisalofthe Intellectualor Mental Capacityofan Individual.
2. Indications ofPossible Personality Disturbance.
Anintelligencetestmay indicate a personality disturbancewhen thetest taker's performanceis erratic. Such

erraticperformance is evidentwhenthe test taker'sresponsesto questions are notconsistent. For example,the testtaker
answersthe easierquestionsincorrectly,but is able to answerthe more difficultquestionscorrectly. Additionally, ifthe
test takergives strange answersto questions, this may be indicative ofa serious mental disturbance.
Intelligence tests cannever beusedalonetodiagnoseapatient,althoughthey cancontributeto theunderstanding

ofthe patient's behavior.
3. IndicationsofSpecial Abilitiesor Limitations
Alow verbal score maybe evidenceoflackofeducationor it may mean the loss orimpairmentofthe power to
use words as symbols as the result ofsome brain injury. Likewise,the test may reveal an unusuallyhigh ability in the
use of vocabulary or arithmetical computation. Depending on the presenting problem to the psychologist, such
information can add to the understandingofthe patient'sskills and potentialities.
4. Observationsofthe Subject's Behavior
Often, personalitytrendsmay bepicked upbyclose observation duringthe testing. Forexample,does the testing
makethe test takerill, does it make him or her tense,or makehim orher at ease? [S. Garfield, ClinicalPsychology113
(1983).]
One ofthe majorproblemswith intelligencetests is that theytend to includean abundanceofverbal testing,and
they also biased towardclasses ofindividualswith particulareducationaland cultural opportunities. Additionally,
are
as stated earlier, if the person administering the test does not follow the standard instructions and methods of
administration designedfor that particular test, then the results will be faulty. Ifthe test is not given and scored as
directed by the test manual, then it cannot be interpreted in terms ofthe existing norms. Even more alarmingis that
2
sometimesthestandard instructions are incomplete orinadequate and the clinicianmust rely onhis orherownjudgment,
thus making the test only as good as the clinicianand his or her judgment. (S. Garfield, 12 at 113).
The two most widely usedintelligence testsare the revised Stanford-J3inet and the variousWechslerScales. (lij
at 121,)
B. Stanford-Binet IntelligenceScale
I. Historyand Construction oithe Stanford-I3inet Test

The Stanford-BinetIntelligence Scaleis designedto measure cognitiveabilitiesin anyone over the age oftwo.
It is used to analyze patternsof thinking, as well as overall cognitivedevelopment. The testconsists ofsubparts that
measuredifferentareas ofcognitive development. These subpartsinclude VerbalReasoning, Quantitative Reasoning,
Abstract/Visual Reasoning,and Short-Term Memory. The tests will produce a Composite StandardAge Scorewhich
isreferredto as "SAS". (}3urosDeskReference,Psychological Assessmentin the Schools(1994), Thetest consists of
15 subsections,whicharenotadministered toeveryone. Theexaminee'sageand perfonnancewill determinethenumber
ofsubteststhat aregiven.
TheStanford-Binetwas originally published in 1916, and revisedin 1937, 1960 and 1972. It is currentlyin its
4tb Edition.Itwas devised as ameasureofchildren'sintelligenceand later revisedto includeadults. (D. Shuman,
note 1, at 51.) The scale is based on mentalage. Mental age is determinedby groupsoftest itemswhich are arranged
jpj
in terms ofage levels, lithepatient is belowtheage ofsix, the itemsare grouped into halfyearlevels (forexample: 4,
4-1/2, 5,5-1/2,etc.). Ifthe patient is six to fourteen years ofage, the items are grouped into year levels of6,7,8, and
so on. After the age of fourteen, there are four levels ranging from "AverageAdult" to "Superior Adult, III." (S.
Garfield, at 113).
At the adult levels, verbal testing is mostly used, as well as problem solving and abstract thinking;whereas,
testing the children'slevel uses blocks and toys, as well as verbal testing.
at (jj
at 122.) Thetestis administeredby
beginningat the level ofthe person'schronological age, orjust belowit, and thenproceedingup a level until the person
fails all the itemson that level, or going down from the initial level ifthe person answers all ofthe items on that level
incorrectly. (jjat 122.) There is no administration time reported for the entire testing, although the Pattern Analysis
subtestdoes havestricttime limits. The test is convenientto use. The flip-overtest booklets havethe directionson the
examinersside, and the display forthe examinee on the other.
The I.Q. (intelligencequotient)is an index whichindicatesthe rate ofmentaldevelopment. I.Q. is determined
twelve, he or she would receive an I.Q. of 120.

approach ofa mean
Tests 2 (1979)]
of 100 and astandard deviation of
(
by dividingthe determinedmental age by the individual's chronologicalage and multiplyingby 100. For example, a
child who is ten and obtains mental age often, would have an l.Q. of IOU. If the same child secureda mental level of
at 123.) Flowever,the Stanford-Binethas adoptedthe deviation
16. [C. Golden, ClinicalInterpretationofObjectivePsychological
Approximatelyfifty percent ofthe peoplewho take the Stanford-Binetobtain a score ofbetween 90 and 110,
twenty-fivepercent achieve above 110, and twenty-fivepercent below 90. Those who score above 140 represent one
percent ofthe population.
The followingshows examplesofwhat is on the test range at the nine year level:
1. PaperCutting — The childisasked to cut out six inch squaresofpaper whichhave been folded
and then is asked to makea drawingofhow the paper would look unfolded.
2. Verbal Absurdities— A series of "foolish" statements are read and the child must tell the
examinerwhat is foolishabout the statement.
3. MemoryforDesigns—A cardwith two designs is shownto the child and then the child isasked
to draw the designsby memory.
4. — The child is to tell the name ofa color, a number, an animal, or a flower which
Rhymes
rhymes with a specific word.
5. Making Change — The child is simply required to make change. For example,1 give you a
dollar for somethingwhichcosts fiftycents. Flow much changedo I get back?
3
6. Repeating FourDigitsReversed—The childmust repeat, backwards, a seriesoffourdigits given
to him or her. (S. Garfield, note 6, at 124.)
On the SuperiorAdult III level, the test taker must be able to define thirty words correctly, such as ocher,
incrustationand perfunctory. (l
at 124.)
The 1960 revision was basedon testing about 4,500 subjects from ages 2-1/2 to 18 years. Only those testing
items whichproved to be the bestwere used in the revision. The reliabilitystandard for the items are generally in the
0.90's. In the 1972 revision, about2,100cases were tested,and thissamplewas evenmore representativeofthe general
population than the original. (aat 125.)
2. Reliability/validity
There are at least two limitations to the Stanford-Binetscale. First ofall, thetest is clearly dominatedby testing
verbal abilities(except at the very earliestages) and is, thus, an inadequate test to test intelligence for a person with
limitedverbalcapabilities,whetherfroma lack ofeducationora differentculturalupbringing. With the growingnumber
((h
ofAsiansandMexicansinour culture, for example,tinsproblemwith thetest shouldbe kept inmind. Secondly, persons
over the ageof 18 were not represented in the standardizedsample, and the testis, therefore, not particularlywell suited
for adults. Additionally, some ofthe actual test itemsrelate mainlyto children or to school. at 125,) Thetestmay
very well be only appropriatefor children from ages 5 to 15 or for suspectedretarded adults. at 126.)
TheStanford-Binet's predictive validity has rangedfromcorrelation coefficients of0.40to 0.75 whencompared
to schoolgrades,teacherratingsandachievementtest scores. The usefulness ofthe test for any other purposeshas never
been adequately demonstrated. (Ziskin, at 215.)
Additionally, evidence indicates that the IQ scores can beaffectedby suchthingsas differentexaminers,home
environment, and the present health of the individual. (1± at 215.) Obviously, a custody fight affects the home
environmentand sometimes eventhehealthofan individual. The clinicianwhohas givena Stanford-Binetin a custody
case probablycannotassurethe court ofits validity,so it is the duty ofthe attorneyoncross-examinationtoexplore this.
C. Wechsler Adult Intelligence Scale — Revised or Third Edition (WAIS—Ri

1. History and Constructionof the Wais-r
TheWechslerscale was developedin 1939, but has been updatedseveral times since then, the latest in 1981.
(kjat 127.) Thistestwas devisedspecificallyto meet theneeds ofa standardized test for adults. The newest version
-
is the Wechsler Adult Intelligence Scale Revised (WAIS-R) and is now considered the most widely used adult
intelligencetest in the UnitedStates. [Lubin,Larson & Matarazzo,"PatternsofPsychologicalTest Usage inthe United
States: 1935- 1982," 39 Am Psychologist 452-453 (1984).]
Not only are the WechslerScales differentfrom the Stanford-Binet in that they were developed for adults, the
WechslerScalesarepoint scalesratherthanmentalage scales. Points are given forcorrectanswers, and then thosepoints
areconvertedintostandardscoresand thento anI.Q. Anotherdifference betweenthe WechslerScalesand theStanford-
Binet is that wherea certaintype oftest item is found throughout the Stanford-I3inet at different levels, the Wechsler
Scalesgroup like items together. Forexample,on the Stanford-Binet, a2-1/2 year old mightbe requestedto repeattwo
digits, and on each new level, the difficulty is increased. On the Wechsler Scales, all of the digit items are grouped
together as one sub-testand administeredfrom the easiest to the most difficult.
TheWechslerscalesalsohaveagenorms foradults. Itappearsfromnormative samplesthat adults reachthepeak
oftheirmental developmentbetweenthe ages of25 and 34. After34, there is a slowdecline. Forexample,according
to theWAIS-Rmanual, a scoreof 115 at ages 25 -34 is equalto an l.Q. of 100. At ages 34 - 44, thesame score is equal
to an I.Q.of 105; and at ages 55-64, it is equivalent to an 1.Q. of 113; and from 70- 74, equal to an I.Q. of122. (5.
Garfield,pm note6, at 127-29.)
The test has two majorcategoriesoftasks: verbal and performance. Eachcategoryhas sub-testswhichmeasure
different abilities, enabling the clinician to analyze the pattem of scores across the sub-tests, as well as verbal
performance and fill scale l.Q.s. This testing method assists the clinicianin the diagnosis ofpsychiatric disorders,
chronic alcoholism, and brain damage. [C. Golden,
4
a. Sub-tests
TheWA!S-Rconsists ofelevensub-tests. Sixsub-tests are intheverbalcategory and fiveare in theperformance

category. Items are arranged in eachsub-test from the simplest to the most difficult. [D. Wechsler,WAIS-RManual:
WechslerAdult Intelligence Scale - Revised(1981).]
Each sub-testappearsbelowin the order in whichitappears in the test. Note that the verbal sub-testsare not all
groupedtogether, nor are the performancesub-tests, but are rather interspersed throughoutthe test:
Information - Verbal
PictureCompletion - Performance
Digit Span - Verbal
PictureArrangement - Performance
Vocabulary- Verbal
Block Design- Performance
Arithmetic- Verbal
Object Assembly- Performance
Comprehension- Verbal
Digit Symbol- Performance
-
Similarities Verbal
b. VERBALSUB-TEST
(a) Information - The information sub-test consists of items to test the test taker's general
knowledgeofinformationofspecific facts. Lowscoreson thissub-testare associatedwith low intelligence. (C. Golden,
2ipmnote90, at 18.)
(b) Digit Span - The digit span sub-test is a test of immediate memory. Scores are given for
memory ofdigits, forwardand backward, fora combined score. (jt at 18.)
(c)
Highscores aremade by those
Vocabulary
with obviously high verbal intelligence and advancededucation. (
- TheVocabulary sub-testisconsideredthebestestimateofaperson's intelligence.
at 19.)
Arithmetic - The arithmeticsub-test,whichis interestingly in the verbal category,measures
(d)
an individual's ability to work number conceptslogicallyand in thecontext ofdaily problems. (j at 16.)
(e) Comprehension - The comprehensionsub-testis a measure of the test taker's ability to
understandsocial customsand to show the appropriate responsein given situations,as well as thereasons forparticular
responses. The sub-testobviouslyassessesthe individual's socialization and assimilation in society, meaninghow well
he relates and interactswith therest ofsociety. The test requires the individual to select,from alternativeanswers,the
most logical answer. Additionally, the test evaluates the test taker's long-term memory and experiences. (l at 15.)
One advantage ofthis particular sub-test is that it elicits remarks from the individualswhich gyreveal a
pathologicalcondition,if the remarks are bizarre or unusuallyodd. For example, ifan individual is asked why we
registermarriages,and he answers,"So that the governmentcan keep track ofchildrenwho are to be sent to top secret
camps," thenthe clinicianwould have a clue that this individual may have a problem. This may even offer diagnostic
clues. (14.at 15-6.) Good scores onthis sub-testis a goodindicationofwhether someoneis in touch with reality. (j4.
at 16.)
(1) Similarities- Thelastsub-test ofthe verbalcategoryis thesimilaritiessub-test. The individual

is askedto tell how two objectsare alike, Forexample,outof"tools"or"clothing"or "horses,"which two are the most
similar? The more abstract the person's association, the higher the score he or she will receiveon the sub-test. More
concrete thinkingis usuallyrelated to the low test scores. (j4.at 17.) Generally,it is thought that schizophrenics tend
to deny thepresence ofsimilarities. (k
at 18.)
c. PERFORMANCESUB-TEST
Performance sub-tests are designedto be more dependenton visual,spatial, or sequentialabilitiesrather than
on verbal skills.
(a) Picture Completion - The picture completionsub-test consists ofa series of drawings or
sketches where an essentialelement is missing. The picture completionsub-test is often the best estimate ofwhat is
called"pre-morbid intelligence," a term used toreferto one'sintelligenceprior to theoccurrenceofaheadinjuryor other
5
mental illness or condition. It appears that paranoid patients often believe that nothing is missing from the picture.
Weiner (1966) believes that schizophrenia can be determinedby what errors are made. For example,a schizophrenic
might finda tongue or sex organsmissing on a picture ofa hot-se with a missing tail. Or, on a violin that is missing
strings, the schizophrenicmight say the violinistis missing. (lit at 20.)
(b) Picture Arrannement - The picture arrangementsub-test is designedto evaluate a person's
perceptionof certain pictures, his or her organization ofseveral pictures, awarenessofappropriatesocial sequences,
(c)
(
planning skills, ability to form and test hypotheses, and flexibility and ability to sequence items in a logical order.
Wechslercalled this test (1958)a measureof"social intelligence." at 21.)
Block Design - The block designsub-testis the purestmeasureofnonverbalreasoningin the
WechslerScale. It is, therefore, the best measure ofnonverbalintelligenceand generalspatialskills. For example,the
skills necessaryin basic engineering may be tested. The test takerviews the block designthat the examinerhas made
andthen attempts to reconstructit. Obviously,the test requiresvisualanalysisskills and visualmotorcoordination. The
test allows the psychologistto observehow the subjectgoes about solvingproblems,whether it is impulsivelyor in an
organized manner.This sub-testisagood measure ofdifferentiatingbetweenschizophrenicsandbrain-damagedpatients.
(C. Golden,2iflU note 14, at 21.)
(d) ObiectAssembly - The objectassemblysub-test is not unlike the block designsub-test in that
Schafterbelieve that depressedindividuals tendto make low scores on this sub-test. (

itrequires visual motor skills. It is unlike theblock design, in that rather than formingabstractdesigns,theindividual
must put togetherpuzzles offamiliarobjects. For a high score, speed and accuracy are essential. Rapaport, Gill &
at 22.)
Digit Symbol - The digit symbol sub-testis sensitive to motorproblemsin the dominanthand
(e)
is requiredas well as quickness. This testis thoughtto be a very good measureofhigh levels ofanxiety. (
and measuresbasic learningskills. The test taker mustalso associate a symbolwith a number. Agood visual memory
at 19-20.)
Lowerscores are often indications of conditions ranging from anxiety to brain damage. This sub-testis well suited to
being given to the culturallydeprived or uneducated becauseit relies very little on verbal skills.
2. IteliabilitylValidityofthe Wais-R
The overall reliabilityofa total score achievedon the WAN is 0.97; however, the WAIS has several sub-tests
which havereliabilityscores as low as 0.60. An expert witnesswho has given the VIAlS may concludethat it may not
reach acceptablereliabilitylimits.
a. Verbal Sub-tests
(a) Information. Although a low information sub-testscoremgi mean lowintelligence, it can also
betheresult ofthe subjectbeing from a differentcultural background than the norm grouporhaving a poor educational
background ormental disorder. (C. Golden, at 14.)
(b) Digit Span. The digit span sub-test,althoughbasically a test ofmemory,can be affectedby a
lack of concentration or attention. Thus, if the test taker is distractedbecause ofthe problems involved in his or her
divorceor custodysuit, then the test resultsmay not be accurate. (Shuman, at 49.)
(c) Vocabulary. The vocabularysub-test,although considered the best estimate ofintelligenceon

the WAIS, is also themost adverselyaffectedby differencesin cultural and socio-economic background. (jd at 49.)
(d) Arithmetic. The biggest flaw with the arithmetic sub-test is that anyonewho is distractedor
highly anxious may have low scores. Once again, so many of the persons who are in custody litigationare naturally
distractedand anxious. Additionally, it is reported that many individualsreact poorly and even refused to answer the
problemspresentedthem becausethey feel the problemsare stupid. Thus, an otherwisebright individual may end up
with a low score. Furthermore,thissub-testmaynotbe a particularlygood indicationofarithmeticskillsdue tothenon-
use ofpenciland paper onthis sub-test; and furtherdue to the emphasisplacedon memory, concentration, attentionand
verbal skills. (C. Golden,jip, at 16-7.)
(e) Comprehension. Whilea good score on the comprehensivesub-test is a good indicationthat
the individual has a good grasp of social realties, studies have shown that the opposite is not necessarilytrue if the
individual has a low score on this sub-test.(l at 16.)
(1) Similarities. Although itis generallythoughtthat in thesimilarities sub-test,schizophrenies deny
thepresence ofsimilarities,it is rare that theanswerson this sub-test are indicative ofa specificpathology. (jjat 18.)
6
Psychological Testing Chapters
b. PerformanceSub-tests
It is generally assumed that the performance sub-tests and the verbal sub-tests measure different things.
However, this assumption appears to be false. Almost every performance test involvesverbal abilities. For example,
theDigit Symbolsub-testincorporates numbersandsymbols,not unlike letters, and it, therefore, offersaclear advantage
to theliterateoreducatedindividual. Thepicturecompletionsub-test,ontheother hand,canonly be understoodthrough
verbal instruction. (1[ at 28-9.)
(a) Picture Completion. There isastrongculturalcomponent to the taskassociatedwith the picture
(
completion sub-tests because familiarity with the object pictured is necessary for high scores on this sub-test. For
example,someonewho has grownup in Harlemand has neverleft mightnotbeabletocompletethe stirrupson a saddle.
at 20.)
(b)
being from a particularculture.
(e)
(
Picture Arrangement. Once again, the biggestdrawbackto this sub-test is the componentof
at 21.)
Block Design. The biggest disadvantage ofthe block designsub-test is the difficultythat low
intelligenceindividuals may have taking it becausethere are not easily answerableitems on this test. (kj, at 21.)
(d) Object Assembly. It has been suggested that depressed individualsdo poorly on the object
assemblysub-test becauseofthe heavyemphasison time bonusesused inscoringthissub-test. (jj
at22.) Theattorney
might want to ask the witness whohas administeredthis test ifsituational depressionin yourclient might affecthis or
her score on this test.
(e) Digit Symbol. Becauseofthe role that visualacuity, motorcoordinationand speed play in this
sub-test, itmay be a more difficult sub-testfor older adults. Studieshaveshown that older adults donot write orhandle
objects asquicklyas younger adults. Thus, there is some question as to whetherspeed should be given any weight in
theevaluationofthe individual's intelligence. [Matarazzo,Wechsler's MeasurementandAppraisalofAdultIntelligence,
215 (1972).]
3. Advantages ofthe Wais-R
There are a numberofadvantages in using the WAIS-Ras a measureofintelligence:
a. The WAIS-Risa standardagainst whichallother I.Q. tests may currentlybemeasured.
b. The WAIS-R does not just give the cliniciana final score, but also sub-test scores,
which can give importantinformation whichan overall score cannot give.
c. It is presentlythemost comprehensive normed adult intelligencetestavailable.
d. The WAIS-R has been the most heavily researchedtest availableto the psychologist.
C. Golden,supra, at 33.)
4. Disadvantagesof the Wais-R
a. The WAIS-R is not suitable for large group testings increases the cost of its
administration.
b. Since the WAJS is biased toward the average American culture group, there are
problemsadministeringit to Blacks,Mexicansand other minoritygroups. (There is an
intelligence test called BITCH,which is the Black Intelligence Test Corrected for
Honkies, which is a test specificallydesigned for black Americans and which is
suggestedbysome researchersto be used in place ofthe WAIS when the individualto
be testedis black).
c. TheWAIShasa tendencytoproducehigherscoreson retesting(as do mostintelligence
tests).
d. TheWAIS has the tendencyto overestimate low I.Q.s. A person with a score of0
might actuallybe an even lower I.Q. (h[ at 33-4.)
e. Themost seriousproblemoftheWAIS is thetendencyofcliniciansto use thetestfor
diagnosticuses, not merely as a tool to measure intelligence. The scientific validity
requires that the test measure what it actually purports to measure. For example,
Wechsler (1944) and Rabin (1941) reported that their studies showed that
schizophrenicsobtain higherscoresontheVerbalScalethan on theperformanceScale.
These findings by Wechslcr and Rabin have not, however, been confirmedby other
studies. Similarcontradictoryfindings ofotherdiagnosticuses oftheWAIS havealso
7
been reported. Many scientific investigators have called attention to the conflicting
results and patterns reported for various clinical populations (Garfield 1948, 1949;
Carter and Bowles, 1948; Guertin, Rabin,Frank and Ladd, 1962; Harper, 1950; Hunt
and Cofer, 1944; Rabin 1945; Rabinand Guertin, 1941). (Garfield, at 130-31.)
Despitethese findings,psychologistshavecontinued to use the WAJSfor diagnosticuses. Since the revision
oftheWAIS, more recentstudiesindicatesomeimprovement inthe quality ofresearchonusing the WAISfordiagnostic
purposes. However, the research still does not revealtruly reliable or valid patternsofopinions on using the WAIS to
diagnose. (Guertin,etal., 1966, 1971). (Jd.at 131.)
G. H. Frank wrote the following,in 1970, concerning the use ofWechsler Scalesas diagnostictools:
"The overviewof the more than twenty-fiveyears of research, therefore, presents us with studies that are
inconsistent, contradictory, and hence, inconclusive. What does seem clear, however, is that the specific
predictions regardingthe performanceofthe suggests in the major diagnosticcategories,viz., schizophrenia,
neurosis, or the brain-disordered,as postulatedby Wechsler and Rapport,have not received support. Indeed,
theresearch si
revealthat there is characteristicpattern ofperformanceon the sub-tests for, for example,
theschizophrenic or theneurotic(page 177). (jj
at 132.)
Any cross-examination of an expert witness who has used Wechsler Scales in his or her evaluation of an
individual should include inquiries into how the WAISwas used in his or her evaluation.
IV. PERSONALITYTESTS
While intelligence tests are designed to evaluate the intellectual ability of an individual, the purpose of
personality tests is to evaluatedifferent aspects ofa person's emotional and social functioning. There are two types of
personality tests - objective and projective. The following covers the major and most used objective and projective
personality tests.
A. Objective Personality Tests
Objective personalitytests present to the subjecttest taker questionsor statementswhich he or she is required
to answerby choosing amonga group ofalternativeanswers. The choiceofanswers may be "true-false," "sometimes-
PersonalityInventory(MMPI). (Shuman, jg
always-never," or "agree-disagree." The most commonly used objective personalitytest is the MinnesotaMultiphasic
note 1, at SI.)
1. Minnesota Multiphasic Personality Inventory(Mmpi)

a. History and Background ofMMPI
(
The MMPI, whichwas developedinthe 1930's and firstpublishedin 1943,has become the "premierdiagnostic
andscreeningdevisein clinical psychology." at 54.) Stark Hathaway,Ph, 11, and John McKinley,M.D.,Ph.D.,the
MMPI's creators, sought to develop a tool to condense psychiatric interviews which were both lengthy and very
expensiveto administerto the mental healthpatient. The MIvIPI'sbasic purposeis that ofa diagnostictool to examine
psychologicalpathology. Much as an x-ray machinelooksinsidethe body,the MMPIseeks insightintoan individual's
psyche. (M. McCurleyand K. Fuller, "The MMPI - What is It?," Advanced Family Law Course, 1986, StateBar of
Texas.)Although psychiatricdiagnosisis farmore difficultthanphysicaldiagnosis,the MMPI has beenusedin so many
settings for over forty years that an enormous body of literaturehas resulted regarding diagnostic and treatment
implications ofvarious MMPI profiles. (14. at 54.)
b. GeneralTerms and Definitions
The following is a listofkey terms used in the application ofthe MMPI.
(1) "MMPI" - As indicatedabove,theinitials standfor Minnesota Personality

Inventory. The testis known as a standardized,objectivetest; however,thereareseveral differing"flavors" ofthe MIvIPI.
The standardMIvIPI most commonlyseen in custody litigationconsistsof566 questions. Some ofthe variations ofthe
standard566question MMPIincludethe MIvIPI-1 68(whichcontainsonly 168 questions),theMini-Mult(71 questions),
theMidi-Mult(86 questions), Fashingbauer'sAbbreviated MMP1 (commonlyknown as the FAM and containing 166
questions), and the Maxi-Mult (94 questions),and the MMPI-2(which will be discussed later in this paper). It should
be noted that currently the various short fonns describedabove aresanctionedby the Universityof Minnesota, which
publishestheMMPIforuse asa researchtool 4y and not for clinicaluse, (McCurleyand Fuller,9iI2 note 40, atT-2.)
8
PsychologicalTesting ChapterS
As seen from the outset, defining the MMPI can be difficult becausethe term may be used generically by
psychologiststo describe any ofthe variations ofthe test. The MMP1 most commonly administeredto individualsfor
individualevaluation, as opposedto group evaluation,is the standard 566 question MMPI. (lcj, at T-2.)
(2) "ITEM" - This is thetermused to referto the questions appearingintheMMP!.

The items are not really questions, however. Each item is an affirmative statementwritten in the firstperson singular,
such as item #231 in the original MM?!, whichreads,"! like to talk about sex.' To each ofthe items, the respondentis
instructed to answer "true"if the statement is "mostlytrue" or "false"ifthe statementis "mostly false," as the statement
appliesto the respondent. If the respondentcannot answer"true" or "false" to the items, the respondentis allowedto
fills in his orher answers on ananswer form.
(3)
(
answer"cannotsay.' Theitemsare presentedto the respondentusually in the formofa test bookletand the respondent
at T-2.)
"SCALES" - The standardMMP! is broken down into !4 categoriesknown
as "scales." Four of the scales are devotedto test validity;i.e., are the responsesofthe test taker answered in such a
mannerthat the testissubjecttoa validinterpretation ofthat person'spersonality?Theremainingtenscalesare "clinical
scales." Theclinicalscales aredesignedtomeasurethepsychologicalinventoryofthe respondent. The standardMMPI,
aspreviouslymentioned,containstenclinicalscales;however, manyversionsoftheMMPIcontainmore thantheoriginal
tenscales. It shouldbe understood at the outsetthat theitemsappearing on the MMPItestform arenot arrangedbyscale.
After the 566 itemswhich makeup the MMPI were selected, thescales were developed by determiningwhich group or
groupingsofitemsand theiranswersindicated certainpersonalitytraits soughttobemeasured. A "Scale" ontheoriginal
MMPI is simplythat group ofitemsand their responses whichpurport to measurea phenomena. For example,part of
Scale 0 (Si) (70 items) is madeup ofitems 32, 67, 82, !! I, I !7, 124 and so on. (Id. at T-3.)
(4) "1" SCORES AND"RAW" SCORES - A "raw" scoreon theMM?! indicates

howmany itemsin theparticularscalethe respondentanswered in the "criticaldirection."The"critical direction"means
the item was answered in such a manner to elevate the score; i.e., ifa respondent answered the item, "! do not like
everyoneIknow,", "false," this would elevate the score onthe "L"scale. The "T"score is the eventual scoreshown on
knowledgeofstatistics. (
theMM?! score sheet. Normal "T" scoresgenerallyrange from 30 to 70 on each scale, The T-score is arrived at by
takingthe raw scores through variousstatistical calculationswhicharenot easilyunderstandablewithoutasophisticated
at T-3.)
(5) MALE/FEMALE SCALES - The designofthe MMPI recognizes that males

and females may, and in some circumstancesshould, respond different!yto certain items appearing on the MMP!.
Therefore,malesand femalesarescoreddifferently, and the scoresheet will usuallyso indicateby denotingMMPI-Male
or MMPI-Female. (hjatT-3.) Aspecia!warningto lawyerson this scale. The Male/FemaleScalesare only a measure
ofthose characteristicsthat havebeenobservedas typical,traditional,conservativemale and femalevalues. So,highly
educatedmales (lawyers, psychologists, doctors, etc.), will score higher on the female scale because oftheirtypical
appreciationofart and other aesthetic values. Likewise,!ower educatedfemales will scorehigheron themasculinescale
becauseoftheirtypicalculturalassociation with moremasculineenvironments(drag races,tractorpulls, etc.). Thus,the
lawyershouldnot betooquick touse this scaletoindict the highlyeducatedmales ashomosexuals orthelower-educated
females as lesbians(which many divorce lawyersare apt to do in cross-examination).This, like all other scales, must
be considered in the socialcontext ofthe subject ofstudy.
2. Construction ofthe MMP!
a. STEP I: SELECTING THE ITEMS: Hathaway and McKinley gatheredover 1,000

questions or itemsthat had !ong beenin use in the psychologicaltexts and psychiatricexamination formsoftheirtime
(circa 1940). After reviewingvarious researchworks and after exercisingtheir subjective and professionalopinions,
Hathaway and McKinley,narrowedthe item pool to550. (Remember, !6 items are repeated.) (Ij at T-4,)
Accordingto the creatorsofthe MM?!, no item was eliminated from the final pool ofitems only because its
manifest contentseemed to have no relationto the psychologicalsyndrome in question. 1-lowever,they claim that no
item was arbitrarily includedin the finalpool ofitemsifthevalidatingevidencefortheitemswas notstrong. Hathaway,
S.R,, Basic Readingson the MMPI in Psychology and Medicine,University ofMinnesotaPress, Minneapolis, 1956,
p. !06. Put anotherway,thecreatorsof the MMPIdid not throwout any itemsmerely becausethewordingoftheitem
did not seem, on it face, to have any relevanceto the particularpersonality trait being measured, For example,one
question,purportedlya potentialindicatorofdepression, reads, "I like to flirt," Although to the laymanthis statement
(
maynot, on its face,seem to haveanythingto do with depression, accordingto therationaleofHathaway and McKinley,
therewas prior validatingevidencethat a fa!se answer to that statementis an indicatorthat aperson may be depressed.
at T-5.)
b. STEP2: GROUPING THE ITEMSINTO GENERALCATEGORIES: As the items
were being selected, they were divided into 26 different categories. These categories includedtopics ranging from
"General Health"to "Habits"to "Obsessive Compulsive" behavior.
9
c. STEP 3: SELECTION OF THE NORMATIVE SAMPLEGROUP: The scoring and
interpretationofthe MMPI is essentiallybased on comparing the score of the individual taking the test with that of a
sample group ofpsychologically nonnal and abnormalpeople; i.e., did the personanswer the questions like the normal
folks or like the "crazy" ones.
Thecharacter and attributesof the normativegroup are very importantbecausethe characterand attributes of
the group comprise the cornerstone of the original MMPI. One researcherdescribed the importance of the MMPI
normative group as follows:
theperformance ofthese menand women on each ofthe scales in the MMPI is used as
thebasis forthenorms in thetest profile. Each subject takingcomponent
theMMPI,therefore,is being compared
to the way a typical man orwomanendorsedthoseitems. In 1940, such a Minnesotanormaladultwas
about thirty-five years old, married, lived in a small town or rural area, had eight years of general
schooling, and worked at a skilled or semi-skilled trade (or was married to a man with such an
occupation level)." Dahlstrom, Welsh, and Dahlstrom, An MMPI Handbook, Vol. I: Clinical
Interpretation. UniversityofMinnesota Press, Minneapolis,1972, pages 7 - 8. Colligan, 1983,
pages67- 68,for adetaileddiscussionregardingthediscrepancies betweenthe original 1940group and
its refined subsethere identifiedas the 1957 group.
d. STEP 4: CREATING THE SCALES: Asstated earlier, theoriginalMMPIhas 14scales
(four validity scales and 10 clinicalscales). A basic understandingofhow thesesscales were developedis essential to
understandingthe MMPI. As discussed above in Step 3, normalsample groups were selected. Additionally,sample
groupsofindividuals diagnosedas havingvariouspsychologicalailments,such as severedepression,hypochondria,and
schizophrenia,were selected. Essentially, the clinicalscales were establishedby givingthe MMPIto a group ofnormal
respondents and a group of, for example,schizophrenics. The scale for schizophrenics would then be developed by
comparing how the two groups answered various items, For example,the scale ofthe original MIvIPI denotedScale 8
(Sc), (schizophrenia was essentiallydevelopedby noting which items the schizophrenicsanswered in a true or false
manner andcomparing thoseresponsesto the responses ofthenormalsampleand groupingthe itemscommonlyendorsed
by the schizophrenicsinto Scale8 (sc). (j.j at T-6.)
3. Reliability and Validity ofthe MMPJ
IftheMMPIwere a perfect instrumentfor inventoryinga person'spsychologicaltraits, one would expect that
unless the individual goes through a majorchangein personality,multipleMMPIsgiventothe individualwould produce
similar results. Research has beendone by RogerGreenattemptingtoexamine"testandretest" reliabilityfor theMMPI.
Additionally, according toreviewersinBurosFifth MentalMeasurementsYearbook,[O.K.Buros,Fifth Mental

MeasurementsYearbook(1959)] the validity for distinguishingone kind ofgroup from another,in terms ofpathology,
is modest at best. As the manualstates, "A high scoreon a scale hasbeen foundto predict positivelythecorresponding
final clinical diagnosis or estimate in more than 60% of new psychiatric admissions." () It does not take a
mathematical geniusto figureout that the MMPI fails to accuratelypredictpersonalitydisorders in almost 40% ofthe
cases. Shouldsuch predictiveerror be allowedin the courtroom?
There is noresearch indicating that the MMPIis 100% valid ortotally invalid. The weightofthe psychological
literatureprobablycould be summed upby sayingthe MMPI is a valuableclinicaltool but shouldnot berelied on to the
exclusion ofeverythingelse.
Considersome ofthe descriptionsquotedby Jay Ziskin in his work, Covingwith Psychiatricand Psychological
Testimony. Ziskindiscussesthe validity and reliabilityofthe MMPI as follows:
"The test is dependentfor its power on self-description. Itwas empiricallydevelopedfrompatient populations
that were reasonably cooperative and reasonably motivated to reveal upset. In a differently motivated
population, the test and its standard norms are not valid and can be grosslymisleading.
Validity has always been held to be the most importantaspect of a test review,yet in this case, it is a most
difficult matter to evaluate. We might well decide that the MMPI as an instrument is valid for many
interpretations and purposesbut a varying levels ofeffectiveness. This would be my position in general. Yet
it would also be possible to doubt the validity or "adequacy"ofa global personalitydescriptionbased on the
MMPIalone. Relativelyfew studieshave madeconcertedattacksontheproblemofglobal validity. These were
encumberedby methodologicalproblems,but, regardless ofthis, resultswere quite disappointing. Using the
devicesofQ sorts and true-falserating scales, testjudgesagreed with therapistsandlor interviewjudges with
nogreater correlation than0.40. Yet, it is with the aspect of"global'validitythat we aremost concerned. It is
a complexquestion." Ziskin, at 220-21.(Emphasisadded.)
10
Although the MMPI is described as a "standardized, objective" test, the terms can be misleading. All that
"standardized" means is that the test asks the same questions in the same order of every person taking the test. All
"objective" meansis that the person taking the test has only three responses: true, false, or cannot say.
Regarding the MMPI as a whole, the test may be a good clinical tool for diagnosingand treating mentalhealth
patients. However, the question arisesas to whetherthe test is a goodindicatorofwho is thebest parent. There are few
items dealingwith "the family"on the MMPI and fewerstill about children. There are no items that say, "1 lovemy
child" or "I'd rathersee my son's baseballgame than play golf.'
While thenewnormative sampledeveloped by Colligan, et al., isaprobableimprovementover theMMPIin that
a newsample group ofnormalswas established,what about the abnormals? The work did not create new psychiatric
samples. It would seem, therefore, that criticism ofthe MM!'! based on the scant informationconcerningsome ofthe
small samples of psychiatricpatients done in the 1940'swould still be valid. Additionally,many ofthe items seem
outdatedin their content.
Further, eventhoughthere is anewnormativesample, thesamplegroup of1,408 people are all formMinnesota.

Colliganreportsthat "withthe possible exceptionoffive subjectsforwhomwehave noinformation,alloursubjectswere
white,' Colligan, 1983, page83. Statisticaltheorynotwithstanding, is it fairto comparethe resultsofthe testingofone
Hispanicfrom Houston to the testimonyof 1,408 whites from Minnesota?
One last general observation concernsdiagnostic languageand categories. Many ofthe abnormalgroupsused
to create the tenclinical scales on the MMPI were diagnosed as havingdisorderswhich have outdated labels. Trying
to findwheretheschizophrenicof 1940 fits into DSM 111 (Diagnosticand StatisticalManual ofMentalDisorders, 3rd
Ed.) is quite difficult, If the definitions have changed, shouldn'tthe clinical scales and their developmentalso be
updated? (McCurleyand Fuller, pm. at T-48.)
4. MMPi-2
a. Differences from the OriginalMMPI.
According to BeverlyKaemmer,a publications representativeofthe UniversityofMinnesotaPress,notmuch

has changedwith the revisions,other than removingitems that were thoughtto be "offensive"suchas questionsabout
religion,sexualpractices, andboweland bladderfunctions. Infact, theitemsthat were deletednever scoredon the basic
scales. A smallpercentage ofthe items were revised to eliminatesexist language, make the questionsmore clear, and
makethecontentmore approachable. Other itemshavebeen designed"to assesssuch behavioras treatment compliance
and amenability to change." (S. Hathawayand J.C. McKinley,"Fact Sheet on the MinnesotaMultiphasicPersonality
Inventory -2 (MMPI-2)", MMPI-2 WORKSHOP,Section 3, UniversityofMinnesota, 1989.) One proposed revision
includes 100 new itemsfor a versionofthe examgeared specificallytowardadolescents— an importantrevision to note
when considered in conjunction with an adolescent'sabilityto file a "ChoiceofManaging Conservator" affidavitwith
the Court.
In arrivingat the changes, a research form ofthe MMP! bookletwas developed,which containednot only the
original 550 items, but an additional 154 items, most ofwhich:
were intendedto replace onesthat were culturallyoutmodedor psychometricallyunsound in the existing
inventory as well as to serve as sourcesofsupplementarymeasuresin the areas offamily dynamics, Type A
behavior, eatingdisorders,substanceabuse,suicide and readinessfor treatmentrehabilitation." (kl)
The experimental booklet,dubbed "AX" as an amalgamfor"Adult Experimental"was conducted as a parallel
study to the adolescentMM!'! study referencedabove. In addition to the experimental booklet(s),subjectswere asked
tocompletespecialsupplementary forms, whichwere designedto assessrecent changesinthesubject'slives, including
(I)
the degree of satisfaction the subjecthad in his or herpersonal relationships.
The samplegroup which took the new experimental versions of the MMPI ranged in age from 16 to 90; the
normative samplegroup waswhittleddown to 1138 males and 1462females. Oneproblem that the committee in charge
ofthere-standardization still has arguablynot surmounted is the factthat the"Hispanic- andAsian-Americansubgroups
are under-represented in the re-standardization sample." (klj
The biggest changewith the 1989update to the MMPI is in the norms, orsamples. Rather than limitingthe
norms to tests conducted in Minnesota,the norm is nowbased on anation-widesample. As to the nation-widesample,
AnneAnastasiwritesthat:
"The two experimental forms ofthe MM!'! were administeredto nationallyrepresentative normative samples,
including approximately 3,000 persons inthe adultsampleand3,000 inthe adolescentsample. The participants
II
comprise randomsamples within communities chosenbecausetheirdemographic characteristics conformclosely

to the 1980 U.S. Census. The total normative sample wasdrawn from several regions of the United States,
including the states ofMinnesota, Ohio, North Carolina, Pennsylvania, Washington,Virginia and California.
The aim was to obtain a sample that is nationally representativeaccording to urban-rural and geographic
residenceand such demographic characteristics as age, sex, educationallevel, and ethnicgroup membership;
preliminaryanalyses indicatea fairlyclose match with the 1980Censusin these demographicvariables." (A.
Anastasi, Psychological Testing,553 (6th Ed. 1988.)
The MIMPI-2 and the MMPJ are very much the same, but the followingrefinements distinguishthe MMPI-2:
(1) Implementationofnationalnorms.
(2) Tscores are basedoneightofthe basicscales,refiningScalesSand0(masculinity/femininity andsocial
introversion), and generatinguniformT scorestoproduce the same type of two-point and three-pointhigh-pointcodes
as have beenused in thepast.
(3) Although the test booklet isabout the same length (567 questions),sexist languageand objectionable
itemshave been modified;the 16 duplicate items have been deleted;and the item order is changed, therebyfacilitating
ascore ofthebasic scales basedon the first 370 items. The remainderofthe itemsprovidesupplementarymaterial,and
althoughsome items were on the original exam, many augmentthe exam.
(4) More subtle indications of personalitywill be assessed by new scales intended to assess protocol
validity, to provide separatemeasurements of masculine and femininegender roles, and to provide diagnosticaids to
respondentsin clinical settings. (Hathaway, pg.)
Ms. Kaemmer,ofthe University ofMinnesota Press,emphasizesthat the changesare notexpectedtohave that
big ofan impact,because they do not impact on thebasic scales.
b. Construction: Modificationsand Continuities- MMPIVersusMMPI-2.
Drs. HathawayandMcKinley boiled downthe initialpool of1000itemsto 550 itemsorquestionsontheoriginal

MMPI. Sixteenofthe itemsare repeated on the originalMMPI as a fail safe device, for a total of566 items. The items
were subdividedinto26 differentcategories on the original MMPI,delineatedbelow:
General Health(9 items)
Generalneurological(19 items)
Cranial Nerves(II items)
Motilityand Coordination (6 items)
Sensibility(5 items)
Vasomotor,trophic, speech,sccretory (10 items)
Cardiorespiratory (5 items)
Gastrointestinal(II items)
Genitourinary(5 items)
Habits (19 items)
Family and marital (26 items)
Occupational(18 items)
Education(12 items)
Sexual attitudes (13 items)
Religiousattitudes (19 items)
Politicalattitudes - law andorder (46 items)
Social attitudes(72 items)
Affect, depressive(32 items)
Affect, manic(24 items)
Obsessive,compulsive(15 items)
Delusions,(31 items)
Phobias (29 items)
Sadistic,masochistic (7 items)
Morale (33 items)
Masculinity- femininity (55 items)
Items toindicatewhetherthe personis tryingtopaint himselfassocially unacceptable (15 items) (M. MeCurley
and K. Fuller,na)
Once the itemswere differentiated into categories,the next issue to resolve was how to score the results--was
the test takercompatible with normal scores or those ofthe abnormal people?
The MMPI-2 contains 567 items, with no repeats.
12
5. DevelopmentofMM?! and MM?! Norms.
Guidelines for how to score results of the MM?! are based on three normative groups: the I-!athaway and
McKinley Groupof 1940, the Hathaway and Briggs Groupof 1957, and the contemporary normative studyofColligan,
Osborne, Swenson and Offord, of 1983. (jjat 7-9.)
(a) The Original NormativeGroup: Hathaway And McKinley, 1940.
As indicated above, the initial samplewas comprisedofaspecifiedclinicalgroupandofa normalcontrolgroup
made upofvisitors to the University ofMinnesotaHospitaland othergroupsfrom the Minnesotaarea, includingpersons
attendingpre-college conferences at the UniversityofMinnesota,and WPA administration workers. The controlgroup
representeda cross section ofboth sexesofthe Minnesotaarea between the ages of16 and 55. (A. Anastasi, at
527.)
(b) The 1957 Group: HathawayAnd Briggs.
This normative groupingwas comprisedofaportionofthe original 1940 group,and consistedof226 males and
315 females. Presumed to be "normal" the range of the 1957 group's responsesto the MMPI was derived from a
comparisonand contrastwith the responses these two groupsofpeople originallygave to the 1940exam. [Hathaway
andBriggs,"SomeNormativeDataontheNewMM!'! Scales," 364-368JournalofClinicalPsycholouv,Vol. 13(1957).]
For a thorough review ofthe difference between the original 1940 normative group and the 1957 normative
subset, theauthorsrecommend furtherreadinginColligan,Osborne,Swensonand Ofl'ord,The MMPI: AContemporary
Normative Study, 12-14,68 (New York: PraegerPublishers, 1983).
(c) The Contemporary NormativeGroup: CoUigan, Osborne,SwensonAndOfford, 1983.
Colliganandhiscolleaguesdevelopeda newrange ofresponsesfor the MMPI. The Colligangroup alsocreated

a new mannerof calculatingT scores. In a nutshell, if a person scores 70 or higher on any scale, that person has
responded differently from 97.7 of the "normal"subjects taking the MMPI. A score of 70 or above on any scale is
thereforeconsidered "clinicallysignificant." [Colligan,Osborne, Swenson & Offord, pjg,
at 71-9!.]
6. Creating the MMPI-2 Scales.

Originally, the MM?! scales were intendedto differentiatebetween what was considered 'normal" and what
reflectedtraditionaldiagnostic categories. Justbecauseyourclient happenstoscore high on theparanoiascale does not
necessarilymeanthat he orshe is paranoid. Therefore,an elevatedscore on theparanoiascale may simply indicatethe
testtaker tends to be distrustful,investigative and curious, [J.T. Kunce and W.P.Anderson, "Perspectives and Uses of
the MM!'! in Non-psychiatric Settings" in P. McReynolds and C.J. Chelune (Eds), Advances in Psychological
Assessment,Vol.6,41-76 (San Francisco: Jossey-Bass,1984).] Furtherresearch has illustratedthata single elevated
score taken out of contextmay not truly reflect the test taker's personality, and that a multidimensional,overlapping
reviewofMMPIscalesisthepreferred approach.The re-standardization committeechargedwith compilingtheMMP!-2
opted for such a multi-dimensional approach.
7. Other Scales.
Since the 1940's over 300 new scales havebeen created, most by independentinvestigatorswho were not privy
to thedevelopmentofthe original MM?!. The new scales arevaried. Many scales were developedby using normal
samples to assesspersonality traitsthat were not related to pathologicalconstraints. As Anne Anastasinotes:
Some scales have subsequently been applied to the test records of the original MM?! normal
standardization sample, thus providingnormative datacomparableto those of the initial dependencyscales.
Examples ofthenewscalesinclude: EgoStrength(Es), Dependency(Dy), Dominance(Do), Prejudice(Pr),and
Social Status(St). Other scales have been developed for highly specialized purposesand are more limited in
their applicability. Still anothergroupingofMMIIitems is representedby the contentscales developedbyJ.S.
Wiggins. In the construction ofthese scales, item clustersbased on a subjectiveclassificationofcontent were
revised and reftned through factoranalytic and internal-consistencyprocedures. The resulting 13 scales have
provedpromising indiagnosisand mayserve as a usefulsupplementin the interpretationoftheoriginalscales."
(A. Anastasi, at 526.)
Frequently scoredsupplementary scales on the MMP!-2 includethe following:
13
(a) ANXIETY (A)
Anxiety isa factordimensionthat emergeswhentheclinicalandvalidityscales are factor analyzed. Highscores
mayindicatemaladjustment, lackofsocial poise,orthatthepersonis inhibited,cautious,distant anduninvolved;ifmale,
high scorers tend to be effeminate,and may become confused and disorganized understress. Low scorers are active,
vigorous,well-spoken, resourceful,power-oriented, abletoleadandevenmanipulateothers, andprefer actiontothought.
[G.S. Welsh, 'Factor DimensionsA and R." In G.S. Welsh and W.G. Dahlstrom(Eds), Basic Readingson the MMPI
in Psychologyand Medicine (Minneapolis: UniversityofMinnesotaPress, 1956).]
(b) REPRESSION GD)
Repression, along with anxiety, is a factordimension that emerges when clinical and validity scales are factor
analyzed. High scorersare considered submissive,internalizing, conventionaland formal. Low scorers are outgoing,
talkative, informal, jolly,impulsive and shrewd. (jj)
(c) MANIFEST ANXIETYSCALE(MAS).
The ManifestAnxiety Scalewas developedto identi&subjectswith high and low drive (i.e., anxiety)in order
to study the effect of the level of the subject's drive on performance. High scorers are predisposed to experience
emotionaldiscomfort in stressfulsituations,jumpy, subject to excessiveperspiration,emphasize the presentmore than
the fixture,perform well on simple tasksand poorly on complextasks. Low scorers tend to remain calm and unruffled
instressfulsituations, and arerelativelyfree ofphysicaland/orsomaticcomplaints. [J.A. Taylor, "A PersonalityScale
ofManifestAnxiety,"48 Journal ofAbnormal and Social Psychology 285-290(1953).]
(d) EGO STRENGTH (Es).
Thepurposeofthe ego strengthsupplementary scale istoassess theresponseofneuroticpatients to individual
psychotherapy. High scorers are tolerant, lackchronicpsychopathology,have a secure sense of reality, and create a
favorablefirst impression. Low scorershave poorpsychological adjustment, are not well equipped to deal with stress,
are more likelyto be diagnosed psychoticthanneurotic,andhavechronic physicalcomplaintsand/orfatigue. [F. Barron,
'An Ego StrengthScale Which Predicts Responseto Psychotherapy" 17 Journal ofConsulting Psychology323-327
(1953).]
(e) DOMINANCE (Do)

The dominancesupplementary scale seeksto identi& people who are dominantin interpersonal
relationships.
High scorers are not readily intimidated,feel safe and secure, are task-oriented,and persevere. Lowscorers are weak
in face to face contacts,unassertive, pessimistic and inefficient. [H.G. Gough, H. McCloslcy and P.E. Meehl, "A
PersonalityScaleforDominance,"46 JournalofAbnormal and Social Psychology360-366, (1951).]
(1) SOCIALRESPONSIBILITY (Re)
Thesocial responsibilityscale evaluateswillingness to accept
responsibility for one's own behavior and sense
ofresponsibilityto the group. High scorersarewilling to acceptresponsibilityfor the consequencesoftheir behavior,
trustworthy,haveadeep concernforethical and moral problems, reject privilegeand favor, and have trust in the world
in general, Low scorers are unwillingto accept responsibility for their ownbehavior, have flexible values,and lack
dependabilityand trustworthiness. [Jj at 73-80.]
(g) COLLEGE MALADJUSTMENT (Mt)
Thisscalediscriminatesbetweenemotionallyadjusted collegefreshmanandthose studentswhoare
emotionally
maladjusted,ineffectual, pessimisticand procrastinators. Low scorers are adjusted, conscientiousand optimistic. JB.
Kleinmuntz,"Identification of MaladjustedCollege Students," 7 Journal ofCounselingPsychology,209-11(1960).]
(h) MACANDREWALCOHOLISM (MAC).
This scaleis designedto identify alcoholicsand individualsprone to alcoholism. scorers are more likely
to be excessivein usingdrugs oralcohol,and may be proneto black-outs. Low scorers areHigh
less likely to use drugs, and
are shy and conventional. [C. MacAndrew, 'The DifferentiationofMale Alcoholic Outpatientsfrom Non-Alcoholic
Psychiatric Outpatients by Means ofthe MIMPI,"26 quarterly Journal ofStudies on Alcohol,238-46(1965).]
(i) OVER-CONTROLLED HOSTILITY (0-H).
14
Thisscaleattempts to distinguishassaultivefrom non-assaultive prisoners. Highscorershave strongemotional

control,and reportfew angry feelings. Low scorers are emotionally immature, irresponsible, and have little emotional
control. [El. Megargee, P.E. Cook & G.A. Mendelssohn, 'Developmentand Validation of an MMPI Scale of
Assaultiveness in Over-controlled individuals,"72 Journal ofAbnormal Psycholozy,5 19-528 (1967).]
8. Myths and Facts AboutScoring The MMPI And The MMPI-2.

(1) MYTH: YOU CAN FAIL THE MM!'!.
FACT: You cannot fail the MM!'!.
A particular score on the MIvIPI does not mean that without question a person is sane, insane, normal or
abnormal. lathe worst casescenarioin a custodydispute,a high orclinicallysignificantscore ontheMMP1 meansthat
thetesttakeranswereda question in a mannerlike that of a particulargroup of "abnormal" subjects or unlike that ofa
particulargroup of"normal"subjects.
Shoulda person score high on Scale6 (Pa) for example,that personis not necessarilyparanoid. Sucha score
would simplyindicatethat the test taker endorseditems similarto those endorsedby a statisticallysignificantnumber
ofparanoidsratherthan those endorsedby a statisticallysignificantnumber ofnormalsubjects.
(2) MYTH: THE MM!'! IS THE FUNCTIONAL EQUIVALENT OF A CLINIC.
FACT: The MM!'! is a clinical tool.
interpretation ofthe MMPI involves creating profilesfrom the T scores achievedby the individualstaking the
test. Thecreatorsdesignedthe MMPIsuch that scores wouldbe achievedby comparingthe interrelationofTscores of
the 10 clinical scales, instead ofby examiningeach scaleindividually. [Graham,The MMPI: A Practical Guide, 18
(New York: Oxford UniversityPress,(1977).]
9. MMPI-2 Profile Scoring

MMPI profiles are createdthroughthe followingsteps:
(I) The raw scores are tabulated and convened into T-scores with "K" correction as appropriate.
Remember, if the "K" scale indicatesthat the respondent was "faking," the T-score from the "K"scale
is used to "correct" T.scores from theclinical scales.
(2) The T-scoresarearranged in sucha mannerthat the highestscore appears first,the nexthighestsecond,
et cetera.
(3) Dependingon the sophisticationofthe clinician, "codetypes"are used to arrive at interpretations.
(4) An interpretation of "code-type" or interrelationof the scales is primarily premised on consulting
prevailing psychologicalliteratureandclinicalexperience. As of1977, therewere over 9,000 booksand
articles published on the development,administration, and interpretation ofthe MMPI. (.ljat 64.)
The MMPI-2 re-standardization weighedtwo possibleapproachesto scoring. The first involveda manipulation
ofnorntalized'F-scores, as advocated by Colligan, Osborne,Swensonand Offord. The re-standardizationcommittee
decided that the manipulation of standardized T-scores would alter the MMPI profile, and further that such a
manipulationwas notjustified with the MMPI.2, as the normativesampleused tojudge everythingwas large. The re-
standardization committee therefore opted for another approach,which involves:
derivinga composite (or avenge)distributionofthe raw scores on the eight basic clinicalscales,
and adjusting the distributionofeach clinical scaleso that it would match the compositedistribution.
The implementation of such a procedure resulted in a set of uniform T-scores that are percentile
equivalent. . ." (Y.S. Ben-Porath,MMP1-2 Consistency Scales , MMPI-2 Workshop,§ 3. Univ. of
Minn. 1989.)
10. ComputerScoring ofthe MM!'! andMMP!-2

TheMMPIcannow bescoredby meansofa computer. The process is relativelysimple. The completed answer
sheet ofan examineeis fed into the computer,usually by opticalscanningequipment. The protocol is scored, and the
appropriateclassificationrules are applied in order to determine the categories for the profile. The computer then
searchesits memoryforinterpretivestatementsappropriatefor the categories and prints out a report. [Graham, g22,
at 185-86.]
15
Soundseasy as 1,2, 3, It is, exceptfor the lawyerwho would like to figureout the basis ofthe clinical opinions
that resultedin an unfavorable MMPI profile for hisTher client or a favorable MMPI profile for the other side. The
problem is nobodyknows who is doing the "Interpretation" within the internal computations of the computerprogram
which generates the final report (except the computer programmers, the computer interpretation services and their
copyright lawyers). With so much research in the field, some conflicting,some consistent, some validated, some not
validated,some reliable and some unreliable, how are we, as lawyers,to intelligentlyexamine the basis of computer
interpretation? Is it Alter's, Barton's Baughman's, Black's, Block's, Boerger's, Butto's, Byrne's, Calvin's, Carkhuffs,
Carson's,Chu's,Comrey's, Cuadra's, Dahlstrom's, Davis',Distlers',Drake's,Duckworth's,Dunbar's,Edwards', Eichman's,
etc. or some combination ofthe above? (The above is a partial list ofresearch material cited as referencesas Graham.
Graham, 1977, page 2 13-14.) (McCurleyand Fuller, at T-42.)
Grahamlistssix computerreportingservicesin hiswork and alsodescribedthevariousservicesoffered by each
ofhis entitles.Graham, 1977. It is also interestingto note that Grahamsent the exact same MMPI test results to all six
of the computer services. Although Graham reported general consistency among the results, there were some
discrepancies. See Graham, 1977, pages 210-11.
Oneclear contradictionwas observedby Grahamin thereport result.s. Four ofthe reportingcompaniesstated
that thesubjectwas "Likelyto reportsomatic complaints that haveno clear organic basis. Graham, 177, page210. One
of the reports madeno statementsregardingsomatic complainings and another stated the subject is "not at all thesort
ofperson that gets bodilysymptomsto symbolizehis emotional conflictswithoutorganiccause." Graham, 1977,page
210 (emphasisadded). In psychologicaljargon, this statementis contradictoryto the other four reports.
In referenceto the validityofcomputergenerated interpretations ofthe MMPI,Ziskinquotesthe opinionofone
researcheras follows:
"At this stage computerized narratives using psychological-test-based information is little more than an art (or
craft) disguised as science. For the most part, the narrative reports are clinical hunches (often many steps
removeddata) whichare automaticallycrankedout by an electronic beast that will, without conscience, weave
a devastating and sometimescontradictorytale about an individual's personalityand problems. The computer
is a generallywilling and efficientservantthat will readily combineand give back scores ofinformation from
its vastmemory. It cares notat all whethertheinformation stored is from astrologycharts, MMPI codebooks,
RorschachIndices, or Somatotypedescriptors ... the "artisan" nature of this endeavorhas been demonstrated,
the "clinical" astuteness is often compelling, but the "science" is often neglected or ofa tertiary consideration
By far the most hauntingproblem and serious shortcoming of the automatedMMPI assessment approach
remainsthat of system validation. Demonstrating the validity ofcomputer-generatednarratives (like that of
demonstratingclinical interpretations generally) is a formidable task." Ziskin, 1981, page 223. (Emphasis
added.)
(McCurleyand Fuller,irna at T-42.)
There are numerous MtvIPl computer reporting services. Perhaps the most common in use by clinical
psychologistsare: NationalComputerSystems;P. 0. Box 1294; Minneapolis,Minnesota 55440; telephone, 612-933-
2800;and theCaIdwell ReportClinicalPsychologicalServices,Inc.; p. o. Box 24624; Los Angeles,California 90024;
telephone, 213-478-3433.
11. Advantages OfThe MM!'! And The MMPI-2.
The MMPI tests are easy to administer,and unlike many intelligencetests that require administration by the
clinician,they can evenbe administeredby a layperson in most circumstances. The MMPI and the MMPI-2 have long
been recognizedas excellentclinicaltools for diagnosing and treating mental health patients. Moreover,the original
MM!'! is the most widely used personalityinventory,and has the added benefit ofyears of empirical researchbeing
generatedto help refine and interpret interpretation of the test. The test has the added benefit of the validity scales,
constructedto catch attemptsto deceive the examiner.
12. Disadvantages Of The MM?! And MMP!-2.
Implicit in the construction of most personality tests is the assumption that an individual's behavior is
characterizedby consistency, regardlessofthe situation. While intellectualfacilitiesare fairly consistentdespite the
situation,an individual's behaviormayshow considerable changefrom onesituationtothe next, "Thus,teststhat identify
individual traits withoutqualifyingthe situationin which they are likely to be manifestare probablymisleadingifnot
invalid." (Shuman, at 62.)
16
Regarding the MMPI and MMPI-2 as a whole, the tests may be a goodclinicaltool for diagnosingand treating
mental health patients. However, when you examine their validity in a custody dispute, their relevance must be
questioned. As was previouslystated,thereare few itemsdealingwith "the family and fewerstill about children. The
MM?! and MMPI-2 WERE NOT DESIGNEDTO FIELP COURTS TO DETERMINE WHO IS THE BETTER
PARENT!
The original MM?! has been heavily criticizedfor its sample group of norms and because some ofthe items
seemedoutdatedintheircontent. (M. McCurleyand M. McCurley,Psychological TestingMarriageDissolutionCourse,
N65-66 May, 1987.) The sample group of norms for the MIVIPI-2 appearsto solve some ofthe problems that have
plaguedthe MMPI. The newsamplegroup consistedof'larger segmentsofthe population ofthe UnitedStates, crossing
culturalbarriers. The items in MMPI-2 haveremainedfairlyuntouchedso the mostofthe criticismofthe original items
remain unchanged.
The last generalobservation concernsdiagnosticlanguageand categories. Many ofthe abnormalgroups used
tocreate the originaltenclinical scaleswere diagnosed ashavingdisorders whichnowhaveoutdatedlabels. The clinical
scales on MMPI-2haveremained virtuallythe same, as in MMPI. Tryingto use DSMHI-R (Diagnosticand Statistical
ManualQfMentalDisorders)inreviewingthe clinicalscales is quitedifficult. Ifthe definitions have changed,shouldn't
the clinicalscales and their developmentalsobe updated? (M. McCurleyand K. Fuller,
13. Cross-examinationofthe MM?! and MMP!-2 Expert
Some suggested lines ofquestioningare as follows:
(1) Was the MM?! in questionscored usingthe updatednorms established by Colligan, in 1983?
(a) Ifyes, is thereresearchvalidatingthe newnorms?
(b) Ifno, pointout criticismsofold norms appearing through Colligan,1983,and Ziskin, 1981.
(c) Ifthe testwas scored bycomputer,does the computerserviceincorporate theresultsofthenew
norms established by Collican, 1983?
(2) Was MMPIscored, whetherby computer or by the clinician, using the newcalculationsregardingT-
scores developed by Colliganin 1983?
(3) How much does the expert rely on the MM?! in the formulation ofhis/heropinion?
(a) Ifthe answeris 100% or "verymuch,"theexpert is in troubleand shouldbe confrontedwith the
research cited in this article.
(b) If the answer is only partly or only as one aspect, the expert is likely making proper use of
MMPI results.
(4) Establish whichMMPI was administered(short form, long form, etc.)
(5) Establish how many and what scales were scored.
(6) Attempt to get theexpert to admit the limits ofthe reliabilityofthe test. Ifthe expert will not concede
the limitations, confront the expert with the Green table cited herein. Also, note Ziskin states,
"Reliability coefficientsfor the MMPIgenerallycluster in the area ofthe 0.70's which is low." Ziskin,
1981, page 217.
(7) Ifthe expertis familiarwith thelie detector testand its reliabilityand validity, have the expert give an
opinionas to whichis a more reliable and valid test as far aspredictionis concerned. Ifthe opinion is
that the liedetectoris morereliableand valid,why shouldMMPIresultsbe admissibleand lie detectors
not. (See Ziskin, 1981, page4, for such a comparison.)
(8) Subpoena the test administeredand the party's answers, Questionssuch as, "So, you're saying just
becauseMrs. Jones says she doesn't have a satisfactory sex life, that means she is a psychopathic
deviant?" should be easily fielded by a competentpsychologist. However, for purposes of cross-
examination ofthe opposingparty, it might be interestingto know andpoint out answersto suchitems
as "Evilspiritspossessme at times."
Additionally,16 ofthe MMPI itemsare repeats. Those itemsare:

1. 8-183;
2. 13-290;
3. 15-314;
4. 16-315;
5. 20-310;
6. 21-308;
7. 22-326;
8. 23-288;
9. 24-333;
10. 32-328;
11. 33-323;
12. 35-331;
17
Psychological Testing Chapter5
13. 37-302;
14. 38-31!;
IS. 305-366; and
16. 317-362.
Collingan, 1983, page67. (It should be noted that the repetition ofthese itemswas done so the MMPI could be
more economically scoredby the IBM 805 machine,not becauseofintentional design. Colligan, 1983,page 67.) Ifthe
answersto the repeatquestions arenot consistent, regardless ofthe scoreson the validityscales, a good argument could
be made against testvalidity.
(9) ifthe MMPI was computerscored, ask iftheexpertknowsthebasis ofthe interpretationplacedon the
scores by the computer; i.e., what research does the computer endorse? Dahistrom's? Green's?
Graham's? Hathaway's? etc.
(10) Vasthetest administered in theexpert'sofficeor was theparty allowedto takethetesthome. Ifthetest
was taken at home, how does the expert know who took the test? (It would be rare that a home test
would be permitted; although, note that in establishingthe new norms, Colliganmailedthe MMP1's to
thesubjectswhotook the test at home. Colligan, 1983, pages 75 and 333. This may be fruit forcross-
examinationitself.)
(II) Ifsub-scaleswereused,were thetest scoresdifferentonthesub-scalesthanthestandardclinicalscales?
Rememberthe sub-scalesare "content" oriented insteadof empirically' developed.
(12) If the expert diagnoses your client as a manic-depressive (or whatever),determine what the expert
observedor wastoldabout yourclient's behaviorandactionswhichsupportsthis diagnosisindependent
oftheMMPIscores. Attack theopinionwith evidenceofopposite behaviorifit exists.
(13) Other Sugeestions, As noted by the researchdone by Green, MMPI scores may vary over time. In a
case where your client has fared poorly on the MMPI, a retest done by a consultingpsychologist may
look substantiallydifferentand provide good information for cross-examination.Note the validating
scales on the MMPI should pick up any deliberate attemptto make the test results look different.
Ina casewhereMMPIresultsarepivota!,sendingtheIvIMPI to the variouscomputerscoringservicesmay(small
chance)result in differentinterpretations ofthe same test. The same couldbe done by taking the test resultsofanother
clinicianand gettinga second opinion. (McCurleyand Fuller,wrnat T-52.)
A valuable source for statistical information and reviews ofMM!'!testvalidityand reliability is Buros,0. K.,
For valuable research information concerningthe reliabilityand validity of MMPIshod form tests, one should
consultGreen,it. L., SomeReflectionsonMM!'! ShortForms: A literatureReviy, Journal ofPersonalityAssessment,
Vol. 46, No, 5, 1982, pages 486 - 487. (Green basically opines the imperfections of the standard MMPI are only
magnified by use ofshort forms.)
B. I6PF
1. History, Background& Purposes of16 PF
Throughtheuse ofa statistical processcalled"FactorAnalysis," the 16 PersonalityFactor Questionnaire (I6PF)
has beendevelopedtoassesspersonality traits. MarcJ. Ackerman,Ph.D.& AndrewW. Kane, Ph.D.,Flow To Examine
Psychological Expertsin Divorceand OtherCivil Actions200 (1990). It is designed for use with individuals that are
sixteen years ofage or older and to analyze suchtraits as "reservedversus outgoing," "humble versus assertive," and
constructionisinadequate.
iiprii at 542-43.
aa
"trusting versus suspicious." These scales are short, the information is on normative samples, and the test
The test never gainedthe acclaimthat the authorshad hoped. a;
see A. Anastasi,
R. B. Cattell, in anefforttoarriveata comprehensivedescription ofpersonality,began assemblingall personality

trait names occurringeither in the dictionary,as compiledby Odberg(1936), or in the psychiatric and psychological
literature. See Anastasi, Cattell developed 171 traitlistswhichhe employed firston a heterogenousgroup of100
adults. Thereafter, he used intercorrelations and factor analysis, and conductedfurthertests on 208 men on a shortened
list. From these factorial analyses, Cattell was able to identify what he described as "the primary source traits of
personalties."
Cattell'sfindings havebeen criticizedas not true identificationsofpersonalitytraits, but instead as reflections
ofthe influenceofsocial stereotypesand other "constanterrorsofjudgment." a in fact, some scientistshave found the
a
same factorswhen analyzing ratings given to completestrangersas when analyzingratings assignedto persons whom
theraters knewwell. (Pessini& Norman,1966). It is arguablethat factoranalysis ofratings may revealmore about
theraters than about the ratees. Id.
18
bestknown is the 16 PersonalityFactorQuestionnaire (Cattell,Eber, & Tatsuoka,1970).

versusventuresome" and ttmstingversussuspicious." j j
Cattell, on the basisofhis factorial research, has constructeda numberofpersonalityinventories,ofwhichthe
Cattell'stest was designed
for ages 16 and over, and yields 16 scores insuch traitsas "reserve versus outgoing," 'humble versus assertive," "shy
A "motivational distortion' orverificationkeyisalsoprovided
for some ofthe forms. A computerized, narrative reporting service is also availablefor users.
Similar inventories havebeen developed for ages 12 to 18 (High School PersonalityQuestionnaire),8 to 12

(Children's PersonalityQuestionnaire), and 6108(EarlySchoolPersonality Questionnaire). is!. Separate inventories have
also been published within more limited areas, including anxiety, depression, and neuroticism. Id. These areas
correspond to certain second-orderfactors identified amongcorrelatedfirst-order factors. j4,
Another addition to the series is the ClinicalAnalysis Questionnaire,a 28-scale inventorywhich includes: a
shortened version of the I6PF "in clinicaldress", with fewer itemsper factor, reworded to fit clinical context; and 12
pathologicalscalesidentifiedthroughfactoranalysisofitems fromthe MMPIandotherclinicalscales. Otheradaptations
ofthe I6PFinspecial dress havebeenpreparedforassessmentpurposein such contextas careerdevelopment, marriage
counseling, and the evaluationofbusinessexecutives.
furtherdevelopment, standardization, and validation.
jj All ofthese inventories are experimental instruments requiring
The assessmentofresponsebias is especiallyimportantin forensicpsychologicalassessments and constitutes

oneofthe unique advantagesofpsychologicaltestingin this field. This is becausein forensic evaluations,by definition,
subjectsgenerallyhavesubstantial, tangiblegains,from eitheraccentuatingtheir strengthsor theirweaknesses. For this
reason,psychologicaltests constructedwith validityscales designedto detect response bias are particularlyuseful.
2. Reliability andValidity cithe I6PF
Reliabilities of factor scores for any single form of the I6PF are generally low. ith at 543. Parallel form
reliabilities centeraround .50 and retestsafter a weekorless often fall below .80. jj Others havequestioned whether
the factorial homogeneityof items within each scale is reliable. (Lovenian, 1961). Id. Available informationon
normativesamples and other aspectsoftest constructionis inadequate,
profiles for more than 50 occupational groups and about the same number
jjof a
Empirical validationdata include average
psychiatricsyndromes. "Specification
equations"are provided for a number ofoccupations, in the form of multiple regressionequations for predicting an
individual's criterionperformancefrom scores on the L6PF. a
In general, validity scales are designed to assess types of exaggerationor minimization in actual clinical
evaluations. However,moststudies ofthe validityscalesofthe I6PF haveusedexperimentalsamples, suchas students
instructed to "fake good" or"fakebad" whiletakingthetest, l3raun &LaFaro, 1969; Riggio,Salinas,&Tucker,
1988;Stricker, 1974;WinderO'DelI &Karson,1975) orjob applicantsundergoingpsychological screenings(Birenbaum,
1986; Birenbaum & Montag, 1989; Elliott, 1976A, 1976B; Kochkin, 1987). Only a few studies existon the use ofthe
I6PF with forensicsamples.(Audubon& Kerwin,1982; Dalby, 1988; Irvine & Gendreau, 1974).
Becauseofthescarcityofresearchontheefficacyofthe I6PF'svalidityscalesinforensicsettings,little is known
about their efficacyindetectingclinicalpatientswhoattemptto "fake good" or "fakebad" (Green, 1988). For example,
there havebeen few studies (Dalby, 1988) that haveassessedthe effectivenessofthe I6PF indetectingresponse bias in
forensicpatientswho are knownto minimize or exaggerate psychopathology on the MinnesotaMultiphasic Personality
Inventory(MMPI) (Hathaway& McKinley, 1967), whichhas the most widely researchedand effectivevalidity scales
ofany psychometric instrument (Greene, 1980, 1988; Ziskin & Faust, 1988).
In addition,the I6PF literaturehas not produceda consensus about thebest cut-offscores to use in deciding
whethera particularprofile shows significantresponse bias (Krug, 1978; Winderet al, 1975). Theoriginalcut-offscore
proposed by Winder et al (1975) was set at seven or higher (out ofa possible 15 items per scale) for both motivation
distortion(fake-good) and the fakebad scale. Krug (1978),using a larger normative samplefor cross-validation, later
suggested that ascoreoften constitutedabettercut-offcriterionforthe fake-good scalebecausethisscoreclassified 15%
ofhis normative sample as a attemptingto fake-good when taking the I6PF under standard instructions. Use ofthis
rationale,however, assumes a normaldistributionofresponsebias in the populationbeing tested. That assumptionis
not a valid one in the populationofparentsbeing evaluatedfor custody.
As Grossman, Haywood& Wasyliw(1992) concludedin their study ofthe use of I6PF validityscales in the
forensicpsychological evaluations ofallegedsex offenders,validityscalesofthe I6PFweresignificantly correlatedwith
those ofthe MMPIand correctlypredicteda highpercentageofpatientswho showed minimization and exaggerationon
theMMPI. Linda S. Grossman,Thomas W. Haywood & Orest E. Wasyliw,The EvaluationofTruthfulness inAlleged
Sex Offenders SelfReports: J6PFandMM?! Validity Scales, JOURNALOF PERSONALITY ASSESSMENT 273
(1992). Theirstudy furtherconcludedthat minimization is far more common amongsex offenders than exaggeration
in comparison to normalpopulations. a
19
In conclusion, factoranalysis providesa technique for grouping personality inventoryitems into relatively
homogeneousand independentclusters. Suchagroupingshould facilitate the investigation ofvalidity againstempirical
criteria and such contribute toward construct definitionand permit a more effective combination of scores for the
prediction of specific criteria. Homogeneity and factorial purity are desirablegoals in test construction,but are not
substitutes forempirical validation.
B. Prolective Personality Tests
Unlikethe "objective" personality tests, projectivetests presentthe testeewith a stimulusandask for a response.
These tests involvemore use ofthe psychologists's judgment than an "objective"test. These type oftests are generally
more stressful to takebecausetheyare less structuredthan, for example,the MMPI. (Shuman, jnm, at 51.)
R. B. Stuart, in describing projectivetests, authorofTrick orTreatment,states, "...theclinicianwho would use
projectivetests must answerMeehl's(1954)question,'Am I doing better than! could by flippingpennies?' An honest
answer to this questionmust be a qualified'No' (page 89)." [R. B. Stuart, Forewordto Trick or Treatment,(1970).] It
is with the above thoughtin mind,that this article explores the disadvantagesof projective tests.
1. RorsehachInkblot Technique
a. Historyand Rackround ofthe Rorschach

The most widely used projective test is the Rorschach. (Ziskin, l22
at 226.) The Rorschach was first
introducedmore than 50 years ago by a Swisspsychiatristnamed HermannRorschach.
Thetestconsistsoften inkblots printed on separate cards. Five of thecards are coloredand therest areblack
and white orshades thereof. Rorschachselectedthese particular cardsafterextensiveclinical research,which led him
to believethat this methodoftestingcould help in personalitystudy. Rorschach'sresultswere basedon test data from
more than 400 subjects. (Garfield, prn. at 170.)
b. Construction, Administration,and Scorine ofthe Rorschach
The testconsistsoftwo mainparts: (I) the free associationperiod,and(2) the "inquiry." In the first part ofthe
test, each ofthe ten cardsare numbered and are shown one at a time in sequential order. Thesubject is asked to tell the
clinicianwhat he or she sees. All responses, as well as the time taken to respond to each card, are recorded by the
examiner. After the free association period is over, the second part ofthe test, the "inquiry," is begun. The subject is
shown the cards once again, but this time is askedbythe clinicianwhichparts ofthe cardhe responded to and what led
him or her to perceivethe blots as he or she did.
The "inquiry"part of the test is used to ascertain two things, "location" (what part of the blotwas used), and
"determination" (whetherthe form is judged good or bad, whether there is movementin the response, whether other
featureswere used, i.e., color, shading,texture,etc.) ofeach response. These two factors, plus some designationofthe
actual contentofthe response,i.e., whetherthe subjectseesan object, animal,or humanbeing,constitute the minimum
scoring ofone response. A fourthscoringelementissometimesadded whena responseis consideredto be popular (P)
or often used. Varioussymbolsareused to score these four elements ofthetest. These combined responsesare then
combined into what is called a "psychogram," which is a summaryofall the scored responses. (kj at 174.)
The scoring ofthis test is obviouslynotan easy matter. The test is, in great part, dependenton the abilities of
the individual examiner. There are several scoring manualsavailable, but the developmentof the Exner system of
administrationand scoring in 1974 has been consideredto have made a substantialimprovementin the reliability and
validity ofthe Rorschachtechnique.
Exnercollectednormative data fromnormalindividualsand variousdiagnosticgroupsfrom a wide range ofage
groups ( 5 - adult). BecauseofExner's system, a cliniciancan comparethe individual who has been tested to normal
people as well as to those with mentaldisorders. (Shuman, at 52.)
c. Reliability& Validityofthe Rorschach
Althoughthe Rorschach is frequentlycriticizedbecauseofproblemswith reliabilityand validity,theRorschach
continuesto be the most widelyused projectivetest. Despite the popularityofthe Rorschach with clinicians,Arthur R.
Jensen, professorofEducationalPsychology and Research Psychologyat the Universityof California,Berkeley, has
written the followingreview ofRorschachreliability:
"Put frankly,the consensusofqualifiedjudgment is that the Rorschachis a very poortest and has no
practicalworth for any ofthe purposesfor which it is recommendedby its devotees... There are now
20
a numberofmethodologicallyand statistically sound and sophisticatedstudies. Even more important
in terms of doing full justice to the Rorschach is that the good research is now being done by the
Rorschachers and projectivetest experts themselves, often with the thu cooperationoftheirclinical
colleagues whoare highly experienced in the use ofprojective techniques. No longercan itbe claimed
that negative findings are the result of bluenose methodologists of statistics and experimental
psychology, applying inappropriate criteria to an instrument for which they have no sympathy nor
clinicalexperiences, nor intuitivefeeling, and no talent. (page 501)
"In addition,the would-be Rorschacher, ifhe is to hold his own amongthe experts, must possess the
kindofgiftssimilartothe literarytalentofanovelistor biographer, combininga perceptiveand intuitive
sensitivity to humanqualitiesand thepower to expressthese perceptionsinsubtle, varied,and complcx
ways. The Rorschach report of an expert is, ifnothing else, a literary work of art. This is the chief
criterion ofexpertnesswith the Rorschach,forthe research has notrevealedany significantdifferences
inreliabilityorvaliditybetweenbeginnersin theRorschachtechniqueandacknowledgedmasters. (page
502)
"Few other tests provide so many opportunitiesfor the multiplicationof error variance as does the
Rorschach... In the typical protocol,most ofthe scoringcategories are usedrelatively infrequentlyso
their reliabilityis practically indeterminant... Most ofthe combinational scores from the Rorschach
consisting oftheratios and differencesamong thevariousprimaryscores(page 504). Scoringreliability
per se has been determinedvery seldom. Reliabilityof scoring depends, to a large extent, upon the
degreeofsimilarityofthe "trainingofthe scorers and has beenreported as ranging from 0.64 to0.91...
The most extensive determinationof retest reliability is that of Epstein and others who gave he
Rorschach to 16 college students,a total of10 timesover a period of5 weeks. The averagereliabilities
for variousresponsecategories ranged from 0.29 to 0.56... Examiner and situationalinfluenceshave
been increasinglyrecognized in recent research as significant contributors to the variance of the
Rorschach scores. The subject-examinerinteraction is certainly one ofthe most important aspects of
the test. The effect of the setting in which the test is taken and the fact that different examiners
consistently elicit different amounts of various score determinants from subjects, should make it
imperative that futureRorschachstudies be basedupona representative samplingofexaminers as well
as ofsubjects...
"Reliability ofinterpretation is, ofcourse,themost important matterofall. Itmay be stated as a general

principlethat the most crucial reliabilityis that ofthe end product ofthe test which, in the case ofthe
Rorschach, usually consists of a verbal description of personalitycharacteristicsbased on a global
evaluation ofall aspectsofthe subject'sprotocol. Contraryto the usual claim ofRorschachers that this
global interpretation is more reliableormore validthanany ofthe elementsupon whichitis based, such
as the scoresand variousderivedcombinations andindices,asystematicsearch ofthe literaturehas not
turnedupa singleinstance wheretheoverall interpretation was more reliablethan the separateelements
entering into it. rorschachtext books havenot presentedany evidenceofsatisfactoryreliabilityofthe
final product ofthe test and the reviewerhas not been ableto find any such evidence in the research
literature... Here aresome typicalexamples ofwhathas been found. Blasinskskyhad 6highlyqualified
Rorschachers rate 40 subjectson 10 personalityitemswhichtheyagreed could be confidentlyassessed
from the Rorschachprotocol. The correlation betweenthejudges was 0.33. Six other cliniciansrating
thesame traitson thebasis ofthecase story abstractsalone, had a correlationof0.31. The interesting
point is that the 10 rated personalityitemswere speciallyselectedas being the kind ofquestionswhich
the Rorschach, and not particularlythe case history, is supposedto be able to answer." (page 505)
(ArthurR. Jensen, in 0. K. l3uros,Sixth Mental Measurements Yearbook, 501 -505.)
Jensencites other studies ofthe Rorschachwhichshow a meanco-efficientof0.30,which is, ofcourse, quite
low.
It is important for theattorneyto be aware ofastudy by Exnerthat some psychologistsmay cite as evidenceof
reliabilityofthe Rorschach. The studyappearedin theJournalofPersonalityAssessmentin 1978and wasentitled,"The
Temporal Stability ofSomeRorschachFeatures." in Exner'sstudy, 100 normalnon-patientswere retested after athree-
year interval. Of 19 variablesevaluatedin this study, only sevenreachedthe minimumcorrelation of0.80. Therefore,
shouldacliniciancitethis studyasprovingthereliabilityoftheRorschach,he orshe shouldbecross-examinedregarding
the fact that themajority ofmeasuresdo not meet thecriteria for reliability. It should also be noted that even Exner
acknowledges that there are a number ofadditionalmeasureswhichwere not even evaluated. (Ziskin, prii, at 241.)
Concerning validity, A. Anastasisums up, for the attorney,what the research has shown is the validity of the
Rorschach:
21
"Nor can any encouragementbe found in empirical studies of Rorschach validity. Despite a
bibliographyofover 2,000publicationsontheRorschach, the vastmajorityofinterpretive relationships
that form the basis of Rorschach scoring have never been empirically validated. The number of
published studies that have failed to demonstrate a significant relation between Rorschach scores,
combination ofscores, or global evaluations and relevant criteria, is truly impressive. The Rorschach
was found to have little or no predictive or concurrentvalidity when checkedagainst such criteria as
psychiatricdiagnosis,responseto psychotherapy, variousdeterminationsofpersonality or intellectual
traits in normal persons,success orfailure inawide varietyofoccupationsinwhichpersonalityqualities
play an important part, and presence ofvarious conflicts, fears, attitudes,or fantasies independently
identifiedin patients. Thosestudies that appeartoprovide positive results have beenshown tocontain
seriousmethodologicaldefects. [A. Anastasi, 12a1
Jensen points out that the validity studiesofthe Rorschach, which show it to be statisticallysignificant,fall in
the generalrange of0.2-0.4 and are so low that one cannotconcludethat the testhas clinicalusefulness. (Ziskin,
at 230.) jjj,
Jensen even suggests that, until proponentsof the Rorschach can substantiate their claims of validity, the
Rorschachshould be abandoned in clinicalpracticeand that studentsof clinical psychology"not be required to waste
their time learningthe technique." (kj at 230.)
It appears obviousthat surely, in more than fifty years of researchon the Rorschach,some positiveresearch
regardingits validityshouldhave beenproducedbynow,and yet is seemsthere has notbeen. Thequestion,then,ofwhy
theRorschachhas so many followers is a mystery,
RoIfA. Peterson,ProfessorofPsychologyat the UniversityofIllinois,assertsthat it is not likelythat Rorschach
devoteeswill discontinueits use; it is, therefore,incumbenton them to provide the predictivevaliditydata which
their interpretations andpredictionsarebased,forwithoutsuchdata,theirconclusionsare merelytheory. upon (jat
235-36.)
d. Advantaaesofthe Rorschach
Althoughdevelopedto analyzebasic personalitystructure,the RorschachInkblot techniquehas also beenused
by the psychologicaland psychiatriccommunity to examine related aspects of personalitysuch as attitudes,motives,
aspirations, and conflicts. The value ofthe Rorschachas an aid in diagnosis is highly controversial, However, its
usefulness maybe,asGarfieldputs it, dependenton"one's philosophyofdiagnosis." (Garfield, at 176.) Garfield,
Exner,and many othersbelievethat the Rorschachhas valueatthe clinicallevel becauseoften the severityofa patient's
problemis notapparenton initial interviews. Theresultsofthe Rorschachmayprovide information foramore complete
evaluation ofthe patient.
e. Cross-Examinationofthe RorschachExpert
Areas of cross-examination, other than the obvious questions regarding the reliability and validity of the
Rorschach,are the following:
(I) Whatever the clinicianproposesas his or her interpretation ofthe subject'sresponsesto the test, he or
she shouldbe askedwhether thereis contradictory psychologicalliteratureor studiesthat challengehis
or her assessment. If the cliniciandenies such studies, then the attorneymust be armed with several
studiesthat contradicthis or her hypothesis. (The research material in this article is good background
for many ofsuch studies.)
(2) Thereare at least five major Rorschach methods orsystems in the UnitedStatesalone. The five systems
- Beck,Hertz, Klopfer,Piotrowsky and Rapaport-Schafer- differ from each otherenormously.
They
(
differ in basic administrativeprocedures,scoring, and interpretive hypothesis. Exner reports that a
surveyshowed that ofthe five systems,roughly 54% ofthe psychologistsprefer the Klopfer System,
and 34% prefer the Beck System. at 237.) On cross-examination,the clinicianshould be asked
which ofthe five major systemshe or she used. Ifthe clinician usedthe Klopfer System, it must be
emphasized thatalmost half ofall cliniciansthat use the Rorschachdo not use that system. In the case
ofall other systems, less than halfuse any one system.
(3) Exner discovered that when a battery oftests is given, the sequence in which the tests are given can
make a difference in the test results. The administration of one test affects the response given in
different tests that follow. Exner cites studies that show, for example, that the incidenceofhuman
content responseson the Rorschachcan alter dependingon the order in which the differenttests are
given. Thus,theattorneyshouldask the expertwitness iftheresults heorshe obtainedmighthavebeen
differentifgiven in a different order. (This is also true ofother tests.) (hj at 238.)
22
(4) It is highly questionable whether the adversary system is conducive to conjecture. Because child
custody issuesinvolvequestionsofapredictivenature, it is evenmorequestionablewhether we should
allowconjecture in thecourtroom. According toExner, childcustody questions are almost impossible
to answer from mostassessmentdata and particularlyfrom Rorschachdata. (k at 240.) The attorney
shouldelicit testimony from the clinicianwho administered the Rorschachif the clinicianrecognizes
Exneras a known expert inthe field ofpsychology. The attorney shouldthen ask the clinicianifhe or
she knows that Exnerbelievesthat the rorschachis or little or no use in child custodydisputes.
It is clear that theRorschach is a highly controversial test, both as to the methodofusing it and even whether
it should be used at all. Given the wide disagreementwithin the professionofpsychologyitself, "...nojudge or jury
should be asked to Believe any conclusions based upon its use." (jj at 242.)
2. Thematic Apperception Test (Tat)
a. History and Background oftheTAT
TheTAT was developed in the 1930's (Murray, 1943)and continuesto be apopular techniqueusedin clinical
assessment. The TAT has been ranked as the fourthmost frequentlyused psychologicaltest. (B. RitzIer, K. Sharkey
and J. Chudy, "A Comprehensive Projective Alternativeto the TAT," JournalofPersonalityAssessment,1980, at 358.)
The purpose of the procedure, when created, was to "evoke fantasies that reveal covert and unconscious
complexes." (jj
at 530.) The test was based on the idea that when facedwith an ambiguous social situation,a person
will expose his or her own personality. The theory,taken further,is that oncethe per so n is explainingthe occurrence,
he or she becomes conscious ofhimselfor herself,and therefore,becomesvulnerableto scrutiny.
ThecreatorsoftheTATcame up with a set ofpictures,that, in most instances,therewasat least one person in
thepicture with whom thesubjectcould, hopefl.illy,relate. Therefore, therewere separate sets ofpictures chosen for
children,males, females,young adultsand older adults.
b. Constructionofthe TAT
StandardTAT instructions stress imaginationand creativity. The test consistsof30 cards whichdepict people
indifferentsituations. A few ofthecards show othertypes ofsceneswithout peoplein them. The examinermay present
thestandardcardsormay,selectively, choose his ownifhe orshe hasaparticularareaonwhichthey wishto focus. The
subjectis then asked to look at each cardand makeup a story about it. (Shuman, at 53.)
In theoriginalstudy done by H. A. Murray,itwas discoveredthat therewere four main sourcesfrom whichthe
stores were drawn: (1) booksand movies;(2) actualevents ofwhich the subjects were aware; (3) experiencesin the
subject'sown life;and (4) the subject's conscious and unconsciousfantasies. (H. Murray,Explorations in Personality
(2d ed. 1947) at 533.)
The first two sources,Murray felt, were those things which had the deepest impression on the patient and,
therefore, comprised a clue to the subject'spersonality. As to actual events or experiences, Murray pointed out that
"everysubjectalmostimmediately projectshisown circumstances,experiences,orpreoccupations."For instance,in one
oftheearly experiments, sixofthe elevencollegemen whotook the testsaidthat the youthin one picture was a student;
whereasnone ofthe twelvenon-college menwho actedas subjects describedhimas such." ía at 533.) To analyze the
subject? possiblefantasies,each story is read and diagnosed separately, and then an attempt is made to find a uniing
theme.
c. Variationsofthe TAT
Most ofthepicturesinthe TAT are indark shadowytones and most ofthe scenes are low-keyedand depressing
situations. Forexample,in one picture,a shortelderlywomanstands with herback turned toatall youngman. Thelatter
is lookingdownwardand with aperplexedexpression, his hatin his hands, Or,inanotherpicture,there isa heavily built
man, naked to the waist with his headhungdownwardand his arms hanging limplyat his side. Some researchersfeel
such somber picturesmay elicitthe negative thoughtsofthe patient.
Therefore,BarryRitzier,Kevin SharkeyandJamesChudyputtogethera newsetofpictures;however,they used

thesame techniques oftestingas used in theoriginalTAT. Theresearchersused picturestakenfrom theFamilyofMan
photo essaycollection publishedby the Museum ofModernArt (1955). The following criteriawas used to select each
picture:
(I) The pictures had to show "potential for eliciting meaningfiilprojective material" (Murray's only
criterion).
23
(2) ofthe pictureshad to include more than one person.
(3) Atleastone-halfofthe pictureshadto showpositiveexpression(e.g.,smiling,dancing,embracing, etc.)
(4) At leastone-half of the pictures had to show activity other than merely sitting or standing. (Ritzier,
Sharkeyand Chudy, at 359.)
The results ofthe test illustratedthat the newtechnique provided a balance ofpositively andnegatively-toned
stories. The researchers felt that they had developed a TAT which covers a more comprehensive range of human
functioningand, therefore,they maymake a fairerassessment for studyingpersonalitycharacteristics. (l
at 361-62.)
d. Reliabilityand Validityofthe TAT
The fundamentalproblemwith the TAT is that it is subject to an even wider range of distorting factors than
objective tests, or even of some other projective tests. For example, examiner bias, particularly in cross-sex
administrations affect the type ofcontentelicited. (hj at 242-43.)
Leonard Eron, Professor ofPsychology, UniversityofIllinois, among others, states that the TAT is not suitable
conviction." (
for providinga profile ofpersonality traits or a reliable measurement ofany one trait. Eron then states, "Nothinghas
appeared in the literature in the last five years which would serve to refute these conclusionswith any degree of
at 243.)
It is extremelyimportant fortheattorneytoknow,whenconfronted with aclinicianwith negativefindings from
z
theTAT abouttheclient,that theTAT is still being publishedwith theoriginal manual (1943),which gives reliability
or validity data. (lj at 243.) Some researchers have suggested that if the TAT were published today, with no
information on reliability, validity and standardization, it would most likely not attain anywhere near its present
popularity. (j at243.)
J. LI. Swartz, whoreviewstheTAT intheEighthMentalMeasurements Yearbook,explainsthat thebody ofTAT
research does not provide any cohesive knowledge regardingthe applicationsofthe test to personality evaluation.
Reliabilityand validity are especiallya problem with the TAT because"...ofthe free nature ofthe response,the high
degree of interaction with situational factors, &eat difficulty in attaining suitable independentcriteria to validate
inferencesagainst and problemsoftrying to developmeaningful quantitative measures." (jj
at 244.) Apparently,no
one has beenable to convince themajorityofexpertsin this field that the above-mentioned problemswith the TAThave
been solved,which raisesquestions like those on the Rorschachabout the usefulnessofa test that has mixed research
resultsafleroverfortyyearsofinvestigation. Additionally, becauseofthemany variationsinscoringand administration,
individual personality traitsaccurately.
e.
(
theTAT is likely to be as ambiguousto theexamineras it is to the subjectsrespondingto it. Thus, theTAT, like the
Rorschach, remainsa test of controversy,with strong doubts as to its reliabilityand validity and its ability to assess
at 244.)
Advantagesofthe TAT
TheTAT canelicitinformation which isusefulto diagnosis andtreatmentplaningwhen it isused in conjunction
with other test data and the history ofthe patient. Because ofthe recent research data regarding the negatively-toned
storiesresultingfrom theselectionofthe picturesby Murray,iftheclinicianusesthe techniquesby Ritzier,Sharkeyand
Chudy,then the result is an even more valid comprehensiveassessment tool forthe clinician.
f. Cross-Examinationofthe Expert
The lineofcross-examinationofthe TAT should,as with theRorschach,be focusedon the reliabilityofthe test.
Theattorney shouldelicit testimony which shows that anyonewho administersthis testin a child custody situationhas
done so when the great weight ofauthorityshows the TAT results to be of little benefit in such situations and, at very
best, full of conjecture.
Additionally, becauseoftheconcernamongcliniciansregardingthepicturesthat composetheTAT, the clinician
should be questioned astowhetherhe orshe took the studies regardingthe negativelytoned picturesinto accountwhen
interpreting the data derived from the TAT. (Shuman, at 53.)
24
3. SentenceCompletion Tests
a. History, Background and Purposes ofSentenceCompletionTests
Another verbal projectivetechnique, sentencecompletion, has been widely employed in both research and
clinical practice. Generally, the opening words for a sentencestem, permit an almost unlimited variety of possible
completions. Examples might be: "My ambition _; Women_; What worries me ; Mymother________
These sentencestems are frequentlyformulated so as to elicit responses relevant to the personality domain under
investigation. The flexibility ofthe sentencecompletion technique represents one of its advantagesfor clinical and
researchpurposes. Nevertheless,some standardized forms have beenpublishedformore general applications.
The use ofthe sentencecompletion technique as a psychological test is considered to have its origin with the
works ofEbbinghaus(1897) in his studies ofmentalabilities. Binet and Simon (1905) found the method to be useful
for measuring intellectual abilitiesand includedsentencestemsas one ofthe tests in their first battery. However,the
sentencecompletionmethod is nowprimarilythoughtofas a technique for assessingpersonalityand attitudes. Foruse
in personality assessment,the test can be tracedto the word associationtechnique,which originated with the work of
Jung in 1904 and 1906. While valuable material can be gatheredthrough the word associationtechnique, it was soon
realizedthat limitationsexistto this method. Longerand possiblymore structuredstimuli and responses are useful for
theinvestigation ofpersonality.
The firstuse ala sentencecompletiontest for personality assessmentis attributedtopsychologistA. F. Payne
in 1928. Othersbeganto usesimilarinstruments, butnotuntilWorld WarII did thesentencecompletionmethodbecome
popular. A needexistedfor a quick,easilyadministeredpsychologicaltest that could begiven to largegroupsofpeople.
The sentence completionmethod met the needs and soonbecamea part ofpsychologicalbatteriesin militarysettings.
It was used in the Air Force as a screeningdevice, but was probablybest known for its use in the Office ofStrategic
Services. In nearlyall settings, a sentencecompletiontestwaspart ofa batteryoftests orpart ala largersetofdataabout
a personandoftenservedasan aid in subsequent interviews. In militaryhospitals,itwas used asa screendevicetohelp
decidewho should be given more thoroughpsychologicaltesting.
Onesuch test, developedtobeused in AirForcehospitals, wasadoptedforcivilianuseafter the War. The Rotter
Incomplete SentencesBlank (RISB) is oneofthe most widelyused sentencecompletiontests in partbecauseit was one
ofthefew whichallows both qualitativeand quantitative assessment ofresponses.
b. TypesofSentenceCompletion Tests
(1) Rotter Incomplete Sentences Blank Test (Risb)

The RotterIncompleteSentencesBlankconsistsof4O sentencestems.Anastasi,pg,at608. Thedirections
to thetest-takerread:
"Completethese sentences to express your real feelings. Try to do every one. Be sure to make a complete
sentence."
Eachcompletionisratedon asevenpointscaleaccordingtothedegreeofadjustmentormaladjustment indicated.

Id. With the aid ofthese specimen responses, fairlyobjective scoring is possible. The sum ofthe individual ratings
providea total adjustmentscore that can be used for screeningpurposes. The response contentcan also be examined
clinicallyfor more specific diagnosticclues.
Rotterintendedtousethe Incomplete SentencesBlankasan alternativetoa lengthystructuredinterview. While
the test was not seen as exposingextremelydeep levels ofpersonality, it was anticipatedthat information not readily
availableduring an interviewwould often be broughtout. By placing some distance between the examinerand the
examineevia use of the test, it allows the examineeto respond more freely than he or she might in a face-to-face
interview. The examinercan take the information at face value, yieldinga quick overviewofsome ofthe issues for a
given individual.
(a) Scoring
The scoring systemsfor all three forms ofthe test are based upon the scaled responsesofcollege freshmenin
the developmentofthe College Formof the RISB. Eachofthe 40 items receives an "adjustmentscore." Each item is
assignedone offourpossiblecodes: (1) Omission(no responseortoo short aresponse to be meaningful); (2) Conflict
response(reflectinghostility or unhappiness); (3) Positive response (reflecting positiveor hopeful attitudes);or (4)
Neutral response (no significantpositive or negative effect). Both "conflict" and "positive"responses are given a
weighingofone to three to reflect the degree ofsentimentexpressed. The manual providesmanuals for each scoring
category,with men's and women'sresponseslistedseparately.
25
(b) Advantages ofthe Test
This ease ofadministration and initialinterpretationare prime reasonsfor the popularityofthe test. The RISB
consists of4O sentencestemswhichexamineesare asked to complete to create statements which reflect their feelings
about themselvesand others. The test can beadministeredto any sizegroup and littleexperienceor trainingis required.
It has been suggested that theRISB is probablymost effective in theearly part ofassessments,to give the examinees
some distance from the examinerto facilitate theirexpressingthemselves more openly.
Most analyses ofthe 1USD are not objective,but subjective,relying on the clinical expertiseofthe evaluator.
Ackerman, irn. at 287. The manual suggeststhat content analysis ofthe RISB be similar to that done with the
TAT, involving the formation of hypotheses about the individual which are to be compared with other data prior to
drawingconclusions.
TheRISB hasbeensuccessful at identifying peoplewho are depressed,who are anxiousand/orwhohavea poor
self image. ith at 288. The RISS has not been found to be especiallyeffective at assessingchanges in psychological
adjustmentover the short term via pre and post test differences.jj, Note that validitydata are largely based on formal
jj
scoring ofthe RISB, while in clinicalpracticeformalscoringis rarely done. This raisesa questionabout the validity
ofa particularinterpretation,since it is based on theskill ofthe clinician doing the interpretation. ilL
(2) The RohdeSentenceCompletion Method

TheRohde Sentence Completion Method(SCM) consistsof 65 itemsplus an open-endedquestion at the end
ofthetest. Rohdetried tochooseeachstem with great care, including comparingthemwith stems from other tests. She
also arranged the order ofthe stemsto try to lead the individualaway from every day life toward the more inaccessible
areas ofpersonality. The open-ended question is a requestto Writebelow anything that seems importantto you." No
formal scoringsystem exists;therefore, eachresponse is to be evaluatedin the contextofpsychodynamicformulations
and categoriesofneed.
(3) The Sacks SentenceCompletionTest

TheSacks Sentence Completion Test (SSCT) consistsof60 itemswhicharerelatively structuredcompared to
most sentencecompletiontest. Sacksalsosuggeststhat aninquirybeconductedat the end to maximizean understanding
oftheresponses. Thetest is organized around 15 attitudes with four stems perattitude-- e.g., attitude toward women,
towardmother,towardcollege,etc. The degreeofdisturbanceregardingeachattitudeis ratedon athree pointscale, from
none to severely disturbed.
(4) The Forer StructuredSentenceCompletion Test

TheForerStructured SentenceCompletion Test(Parer) consistsof100 itemswith instructionswhichemphasize
speed. The stemsarestructuredtotry toforce therespondenttooffer materialuseful for diagnosis,e.g., to indicatewhen
he orshe was most depressed. Thereareseparate male and femaleforms. Whilethe sentencesarenot formallyscored,
Forer suggestsorganizingthem into sevenmajor areas, and orders a checklist for that purpose.
(5) The Miale-holsopple SentenceCompletion Test
The Miale-Holsopple Sentence Completion Testconsistsof73itemsand is designedto beminimallystructured.

The stems consist ofrelatively general items, rather than obviouslypersonal ones, with the intent ofminimizingthe
degree to which people feel threatenedor exposed. The instructionsare also more general than most, asking the
individual to "complete each sentence in whatever way you wish." There is no formal scoring system. Instead, the
authors urge the clinicianto form hypothesessentenceby sentence,and to combinethem into a global descriptionafter
all sentencesareread. To facilitate identification ofunusual responses, numerous examples ofcharacteristicresponses
are given for each stem. 'Overall, the Miale-Holsoppleis the least structuredof the sentence completion tests and
dependson the subjectiveinterpretiveskills ofthe clinician more than any ofthe other tests.
(6) IncompleteSentencesTask
A more recently developedinstrumentis the IncompleteSentencesTask. Anastasi, at 600. In its
development, this instrument combined the projective approach with standard psychometric procedures of test
constructionand evaluation. Availablein a School Form(GradesSeventoTwelve) and a CollegeForm, the test yields
three scores: hostility,anxiety, and dependence. These constructs were chosen on the basis ofpersonalitytheory and
becauseoftheir clinical importance to any adjustmentofstudent.
As apsychometric instrument, thistestwould be strengthenedwith more representativenorms and with further
researchon reliability, validityand interrelationofthe three scores. Initial resultsare promising,however, and provide
more objectivedata thanare customarilyavailablefor projectiveinstruments.
26
c. Acceptanceby Mental Health Professionals

Inareport presented at themeetingoftheAmerican Psychological Associationin 1983, psychologistB. Luben,
R. Larsenand J. Matarazzoindicated that "Sentence completiontests(all kinds) rank 7.5 among psychologicaltests in
their surveyof psychologists.' The RISB by itself rankedtwelfth and is the only sentence completion test mentioned
by name in their survey. Sentencecompletiontests also ranked7.5 in a 1982 survey,up from 8.5 in a 1969 survey. The
RISI3 also rankedtwelfthin a 1982 survey,down from 10th in a 1969 survey but up from 61st in a 1959survey. [The
MMPIwas the numberone ranked test in this survey ofpsychologists.]
4. Projective Drawin2s
Most psychologistsrecognize that not all the information desired in psychologicalinvestigationscan be gained
throughverbalinteraction. Ackerman, at248. Children start drawingbefore they learn to talk, thereforedrawing
techniques can be veryuseful in assessing a child's intellectual and personal functioning. Thefollowingtests have
been used in the areaofprojectivedrawing and could be encountered in divorce litigation.
a. Typesof ProjectiveDrawing Tests
(1) DRAW-A-MANTEST
Goodenough developed the Draw-a-Mantestin 1926 forchildren betweenthe ages ofthreeand fifteen asaquick
estimate ofthe child's intellectual development. !4
at 249. Goodenough developed a SQ-item scoringscale. Id.
(2) THE GOODENOUGH-HA.RRIS DRAWING TEST

In 1963, Harris updated the Goodenough Draw-a-Mantest to includea drawingofthe selfandofa woman.Id. Scoring
jj
j
criteria were presented for each ofthe drawings. Harris added 22 items to Goodenough's original criteria. Sattler,
one of the foremost experts in intelligence testing, points out that the "Goodenoughdrawing test is an acceptable
screening instrumentfor use asa non-verbalmeasure ofcognitiveability,particularly with children under 10or 11 years
ofage." Sattler felt howeverthat the drawing test should not be used as the only measure ofintelligence due to
reduced validity. a
(3) HOUSE-TREE-PERSONDRAWING TEST
In working with non-compliantchildren, Buck, out ofdesperation, requested subjectsto draw a house, a tree,
and a person. a He chose these three items, as they represented the items most frequently spontaneously drawn by
children.Id. The subjectis asked to draw ahouse, a tree,and apersononseparatepages. After the drawingofthe person
is completed,the subject is asked to draw a person of the oppositesex. jj Although a thorough discussionofthe
interpretationofthe house-tree-person is not possible in a manuscriptofthis nature, "the houseis thoughttorepresent
the environment, the tree depictsgrowth,and the personrepresents the integrationofthesubject'spersonality." ant250.
(4) DRAW-A-FAMILYTEST
TheDraw-a-Family test was originallydevelopedby Appel and later elaboratedon by Wolff. Id. There are
a
j
generallytwoacceptable methods ofpresentingthis test. Onemethodrequiresthe subject to "draw apicture ofyour
whole family, " and the other requests the subjectto "draw a picture ofa family." Someexaminers preferthe more
ambiguous request to draw family as it provides the examiner with the opportunity to measure the subject's
identification with their own family. Family constellation,placementof individuals in the family, relative sizes of
the family members and othervariablesare used in interpretingthe draw-a-family test.
(5) KINETIC FAMILY DRAWING

TheKinetic Family Drawings(ICFD)test wasdeveloped by Bums & Kaufman in 1970, Id. "The approach of
usingkinetic (action)instructions- i.e., askingthe child to produceadrawing offiguresmovingordoing somethinghas
been found to producemuch more valid and dynamic material in the attempt to understandthe psychopathologyof
children in a family setting."j[
Theinstructionsgiven by the examinerstate,"I would like youtodrawa picture ofeverybodyin your family
doing something." a Ofparticular note in the interpretationofthis picture is whether the family is doing something
together, or each individual is doing somethingseparately. a
27
In the context ofa custodyevaluation, the KFD can be very enlighteningas a way ofunderstandingthe child's
perception ofhis family. Evaluatorsusuallyinterpret both thegeneral themes and dynamics represented,and sometimes
the difference in physical distance betweenthe childand eachparent as an indicationofrelative emotionalcloseness or
distance. Somerecent researchhas begunto provideevidenceofa correlation between the distancebetween figures on
the KFD and the child's perceived closeness to the people representedin the drawing,but strong confirmationofeven
that interpretation ofthe KFD awaits the accumulation ofmore data.
As with any psychologicaltest used in the context ofa custody evaluation,each of these four tests has some
utility. Themost reasonableuse ofthese tests'resultsistoserve as one additional sourceofinformationto be considered
in the contextofall the factsin makingrecommendations regardingcustodyand access. Even the Bricklin,the only test
ofthe groupspecificallydesigned for custodyevaluations,doesnot haveenough validationdata to justify its use as the
only test given in a particular case.
(6) OTHER DRAWING TECHNIQUES
There are many minor drawingtechniquesthat have been developed for usc in interpreting specific kinds of
personality concerns. jj
The number of drawing techniques developedonly by the creativity of the examiner. Id.
Examples include, (1) The Draw-a-Person-in-the-Rain Test to measure how the individual deals with unpleasant
environmental stress; (2) The Draw-a-DreamTest wasdesignedto depicta dream that the child actuallyhas had or one
that the child would like to have; (3) The Representational Family Drawingwhich was developedby Oaklanderin an
effort to measurehow the childrepresentationallyperceiveseach familymember.Id. at 251. The instructionof"draw
a picture ofsomethingthat represents each memberofyourfamily" is given. Id.
b. Cross-Examination of the ProiectiveTest Expert
Projective tests,becauseoftheirsubjectiveinterpretation, are topics forintensecross-examination.ifprojective
tests are the only tests administered, expose to thejury insufficient dataexists to base his conclusions.
Some ofthe questions which mightaid the attorney in cross-examinationofthe psychologicalexpert, are as
follows:
(I) Were projectivetechniquesadministered?
(2) Ifthepsychologistused only projectivemethods for testing,why?
(3) Were the instructionsfor the specific projective tests followed according to the manual? Ask the
examinerto repeat the instructions that were given.
(4) Were the projectiveinstrumentsadministeredprior to objective instructions? Ifnot, why not?
(5) What projectivedrawingswere used?
Lynn Little St. Leger,PsychologicalExaminations,AdvancedFamily Law CourseBB, 99-100(1990).

c. Reliabilityand Validity
In general, projectiveinstrumentsareless suspectableto fakingthan are self-reportinventories. Anastasi,
at 613. The purposeofprojectivetechniquesis usually disguised. jj
Even ifan individualhas some psychological
sophisticationand is familiarwith the generalnature ofaparticularinstrumentsuchas the RorschachorTAT, it is still
unlikely that he or she can predict the intricate ways in which these responses will be scored and interpreted. Id.
Moreover,the respondentsoon becomes absorbedin a task and is less likely to resort to the customarydisguisesand
a
restraintsofinterpersonal communication,
On the other hand, it cannot be assumedthat projective tests are completely immune to faking. j
Several
experimentswith theRorschach,TAT,and otherprojectiveinstruments haveshownthat significantdifferencesdooccur
whenrespondents are instructed to alter theirresponsesso astocreate favorableorunfavorableimpressions orwhen they
aregiven statementssuggestingthat certaintypes ofresponsesaremore desirable. Jj
lit is obviousthat most projectivetechniquesareinadequately standardizedwith respectto both administration
jj
and scoring. at 614. Yet evidence exists that even subtle differencesin the phrasingof verbal instructionsand an
examiner-examinee relationships can appreciablyalter performanceon these tests. Even whenemployingidentical
instructions, someexaminersmay be more encouragingor reassuring,others more threatening, owing to theirgeneral
manner and appearance. jj
Such differences may affect response productivity, defensiveness, stereotyping,
imaginativeness, andotherbasicperformancecharacteristics. Id. Inthe lightofthesefindings, problems ofadministration
and testing conditions assume even greater importancethan in other psychologicaltests.] Id.
Equally serious is the lack of objectivity in scoring. jj

Even when objective scoring systems have been
developed,the final steps in the evaluation and integration of the raw data usually dcpend on the skill and clinical
experienceofthe examiner. jj
Butperhaps the most disturbing implicationis that the interpretation ofscores is often
28
as projectivefor the examineras the teststimuli are for the exatninee. aIn otherwords, the final interpretationof
projective test responses may reveal more about the theoretical orientation, favored hypothesis, and personality
idiosyncrasiesofthe examiner than it does about the examinee'spersonality dynamics. jjj
Anothercommondeficiency to many projectiveinstruments pertainsto normative data.
completely lacking, grossly inadequate, or based on vaguely describedpopulations. a jj
Such data may be
In theabsence of adequate
objective norms, the clinician falls back on his or her "general clinical experience" to interpret projective test
performance. jj But such a frame of reference is subject to all the distortionsofmemory that reflect theoretical bias,
preconceptions,and other idiosyncrasies ofthe clinician.Id. Theclinicianmay,thus,lacksufficientfirst-hand familiarity
with the characteristicreactionsofnormal people.
In view ofthe relativelyunder-standardized scoringproceduresand the inadequacies ofnormative data, score
jj
on thebasisoftheirinterpretations ofthe identicalrecord.
been conductedwith projective tests. a j
reliability becomes an important consideration in projective testing. at 615. Interpretative score reliability is
concemedwith the extent to whichdifferentexaminersattribute the same personality characteristicsto the respondent
Few adequatestudiesofthistype ofscore reliabilityhave
Some investigationshave revealed marked divergencesin the interpretations
given by reasonably well qualifiedtest users. A fundamentalambiguity in such results stems from the unknown
contributionofthe interprete?sskill. Neitherhighnor low scorereliabilitycanbe directlygeneralizedtoother scores
differing appreciablyfrom thoseutilized in the particular investigation. a
a a
Retest reliabilityalso presents special problems. at 616. With longintervals, genuine personalitychanges
With shortintervals,a retestmayshowno morethanrecall of original
may occur which the test should detect.
responses. a a
A large chanceof variations are to be expectedunder such circumstances. Ratiosand percentages
computed with suchunreliablemeasures are often evenmore unstablethan the individual measuresthemselves. For a
any test, the most fundamentalquestion is that ofvalidity.j4. Many validationstudies ofprojective tests have been
aa
concemedwith concurrent criterionrelated validation. Mostofthosehave compared theperformanceofcontrasted
both. )
groups,such as occupational or diagnostic groups. The large majority ofpublishedvalidationstudies on projective
techniquesare inconclusive becauseofproceduraldeficienciesineither experimental controlsor statisticalanalyses, or
a at 617. (This is especiallytrue of studies concerned with the Rorschachtest.
deficiencies may have the effect of producing spurious evidence of validity where none actually exists.
customarycontrolforthetypesofcontaminationinvalidationstudiesis toutilize blind analysis, in whichthe testworker
a
Some methodological
The
is interpreted bya scorerwhohashad no contactwith therespondentand who hasnoinformationabout herorhimother

than that contained in the testprotocol. Commissions have argued,however, that blind analysis is an unnatural wayto
interpretprojectivetest responses and does notcorrespondto the ways theseinstrumentsare used in clinicalpractice. a
a
Besidestheirquestionabletheoreticalrationale, many projectivetechniquesare found clearly wanting when
evaluatedin accordancewith test standards. at 621. This apparentcontradictioncan perhaps be understoodifwe
recognizethat, with a few exceptions, projectivetechniquesare nottrulytests. Ratherthanbeingregarded andevaluated
as psychometric instruments, or tests in thestrictsense ofthe term,projectiveteclmiques are coming more and more to
be regardedas clinicaltools. Id. Thus, theymay serve as supplementaryqualitativeinterviewingaidsin thehands ofa
skilled clinician. Theirvalue as clinical tools is proportional to the skill ofthe clinicianand, hence, cannot be assessed
independentlyof the individual clinician using them.
psychometric procedures would be inappropriate. But
a at 622. Attempts to evaluate them in terms of the usual
by same token,theuseofelaborate scoringsystemsthat yield
the
quantitativescores is not only wasteful but also misleading. j4 Such scoring procedureslend the scores an illusory
semblanceofobjectivityand maycreatethe unwarrantedimpressionthat the given techniquecan be treatedasa test,
The special valuethat projectivetechniquesmay haveis more likely to emergewhen theyare interpretedby qualitative
a
clinical procedures than whenthey are quantitatively scoredand interpretedas psychometric instruments, a
V. NEW DEVELOPMENTSIN PSYCHOLOGICAL TESTING WiTHIN THE LAST DECADE
A. Bricklin Perceptual Scales (Bps)
1. History,Backgroundand Purpose ofBricklin
The Bricklin PerceptualScales, developedbyBarryBricklin,Ph.D.,were firstpublishedin 1984. Althoughonly
recently published, the research, testing and validationof the test has been an ongoing process since the early 1960's.
The test was designed to measure "a child's unconscious or nonverbal perception of each parent in the areas of
competence, supportiveness, follow-up consistency, andpossessionofadmirabletraits."(H. Bricklin,BricklinPerceptual
Scales Manual, at 6.)
Dr. Hricklin developedthe test because,althoughhe is an "expert' in the use ofthe standardprojectivetests, he
believed that these tests were not well-suited for use in custody disputes. Bricklin concluded that the following
characteristics were required ofany test which was to be useful to a court in custody disputes:
29
(I) The test would have to be based on the child'sperceptions, rather than on parental behavior.
(2) The test would have to look at the child'sobservations, thoughts and feelingsabout cactiparent.
(3) The test would have to rate each parent in each area coveredby the test.
(4) The test would have to avoid asking the childdirectquestions.
(5) The test would have to avoid askingthe childto makea directchoicebetween parents.
The testwouldhave to be capable ofreflectinga particularneed ofa childthat is so compellingthat the
(6)
(7) The test would havetobe basedprimarilyon unconscious, rather thanconscious,responses.
13.)
(
parent who could best satis& that need would receive recognition ofit in the scoring scheme.
at 12-
Forvalidand reliableuse, Bricklinsuggests that the test be used only on childrenages six years and up.
2. Construction ofBPS
Fourareas are tested on the BPS: (I) the child'sperceptionofa parent's competence, (2) supportiveness, (3)
follow-upconsistency, and (4) possession ofadmirabletraits(e.g., keeping promises,trustworthiness,altruism,ability
to acceptcriticism, etc.),ía at 17.)
Thetest consistsof64 questions,32 of whichpertain to the child's perceptionsofm other and 32 to the child's
perceptionsoffather. First,the child is asked to respondto the questions verbally,and theanswer is recorded. The child
is then askedto respondto the same questions non-verbally on a continuumline which ranges from Very Well to Not
So Well. (j at 17.)
Each question ispresentedby showingthe childa card, whichpicturesa line with the words "Very Well" at the
extremeright and "Not So Well"to the left. (A few cards say 'Very Oftent' and "Not So Often"). (hj at 18.)
Afterthe childhas beenasked the questionsverbally and they havebeen recorded,the nonverbalportionofthe
test begins. Thecard isheldin front ofthechild andclinicianasks, "Ifthis (pointingto the end ofthe line marked"Very
Well") is Dad doing very well and helpingyou calm down, and this (pointing to the end marked "NotSo Well") is Dad
doing not so wellathelpingyou calm down, where on this line would Dad be?" The child then uses a stylusto punch
his orherresponsealong thecontinuum. (jjat 18.) The questions, aswell asthe scoringgrid, are printedonthe reverse
side of the card. The grid ranges from I to 60, with I being "not so well" and 60 being "very well."
It is assumed that thechild's placementon the continuum represents an unconscious,gut-levelresponsewhich
reflects a combination ofobservations, thoughtsand feelings"weighted" by the child'sown value system ofthe parent
being consideredin the area coveredby the card. ía at 19.)
The following are examplesofthe questionspresentedto the child in the BPS:
(a) COMPETENCY
Verbal:
"Ifyou had questions bout wherebabiescomefrom (childrenunder 7and older)sexandbabies (children
8 and older) howhelpfuldo you think you would find it to talk to Momabout this?"
Non-Verbal:
"Ifthis is Mom being very helpful at answeringyour questions,and this is Mom being not so helpful
at answeringyour questions,where on this line would Mombe?"
(b) SUPPORTIVENESS
Verbal:
"IfDadhappensto be theone to take you to the doctor,how is he at makingyou feel comfortableabout
this?"
Non-Verbal:
"Ifthis is Dad doingvery well at helping you to feel comfortable,and this is Daddoing not so well at
helping you to feel comfortable, where on this line would Dad be?"
(e) FOLLOWUP CONSISTENCY
Verbal:
"I'low often does Mom make sure you do yourjobs around the house?"
30
Non-Verbal:
"Ifthis is Momveiy often making sure you do yourjobs aroundthehouse, and thisis Momnot sooften
making sure you do yourjobsaroundthe house,whereon this line would Mom be?"
(d) ADMIRABLE TRAITS
Verbal:
"When Dad has a chanceto spend sometime with other people,how much does he seem to enjoy this?
Non-Verbal:
"Ifthisis Dad very much enjoying time with other people, and this is Dad not so much enjoyingtime with other
people,where on this line would Dad be?" Bricklin PerceptualScales, Cards 10, 17,41 and 56.
3. Supplements to the BPS

To date, Dr. Bricklin has updatedthe BPS with six short supplements,whichare deliveredalong with thetest.
A briefsynopsisofthe supplements follow:
(a) Suspect Reactions

Wherethechildmaypreferoneparentascaretakerover theother,but the evaluatorfeelsthat thepreferencemay
not be in the child's best interest, Bricklin admonishes the test giver to be wary ofsuspect reactions. Bricklin outlines
the following as suspect regardinga child'sconscious or verbal choice:
a. it is based on what the child has been told ratherthan actual experience.
b. the evaluatorsuspectsthat the childis tryingto "save' a parent,for fear ofthat parent falling apart ifnot
chosen.
c. the evaluatorsenses the possibility ofthe child having been manipulated, through gifts, promises, or
special treatment (i.e., lack of rules, letting the child stay up past bedtime, etc.). BPS Manual
SupplementNo. 1 6/11/85.
(b) CloseCalls
Supplement 2 helpspreparetheevaluatorfor cross-examination on close calls,where the contestants havecome

out fairlyequally on test results. To try to furtherassess close calls, Dr. Bricklinsuggests that the evaluatorfollowup
with cards 4, 6BM, 7BM, 7W, and 10 of the Thematic ApperceptionTest (TAT). BPS Manual SupplementNo. 2
9/12/85.
(c) Reports
Bricklin bases his own reports on a combination of the BPS, the PerceptionofRelationshipsTest (commonly
referred to as the PORT), it involved drawing tests), My Parent Would, and the House-Tree-Person Test, He also
sometimesuses the Rorschach. When facedwith a real adversary case,where one sidewants to litigateregardlessofthe
data, Bricklin advocates writinga short report, tying down the pertinent date to prevent "fishingexpeditions." BPS
Manual SupplementNo. 3 3/28/86.
(d) AdditionalDirectionsfor Child With ConsciousBias

Inthemostrecentsupplement,Bricklinaddressesthe "MMU." Ifachildhas a conscious bias, or is a child with
"mind-made-up" (MMU),that child has probablybeenprogrammed orbribed. Bricklinsuggeststhe followingas tools
to ensureaccurateBPS data with MMU children:
a. Speak very slowly when you are reading the questions to ensure that the child understands and
occasionallyask the child to repeatwhat was said.
b. Challengesomeresponses,especiallythosegiven in haste. Ask for examples. Ifthe child cannot give
any exampleto backup the quick response,the response may have been contrived.
c. Repeat critical wordsin long questions.
d. Watch involuntaryhand movements - does the child have an initial impulse in one direction on the
continuumand then force himselfto respondon the other end ofthe continuum?
e. Does the child avoid eye contact? Resist humor or friendliness? BPS Manual SupplementNo. 6
6/30/88.
31
4. Scoring the BPS
When 64 cards are completed,there will be 32 paired scores, 11 ofwhich measure Competency, 11 measure
Supportiveness, 3 measure Follow-UpConsistency and 7 measure Admirable Traits.
The cliniciangoes throughthe paired Point Scores foreach item and draws a circle around the highernumber.
The clinicianthen counts the numberof items on whichDad had the highestscore and the number ofitems on which
Momhad the highest score. The parent achieving the higher number ofhighest scores perceives that parent as the
primarycaretakingparent. (lc at 33 -4.)
5. ReliabilityandValidity ofthe BPS
Bricklin, in dealingwith reliability, notes that "there are no reasons to expectthe measurements reported here
to exhibitany particulardegreeofstability,sincethey shouldvaryinaccordancewith changes inthechild'sperceptions."
Ackerman, at 282, citingJ3ricklin PerceptualScales,at 36. Limitedreliabilitystudies havebeen conducted, one
in particular, where 12 children incustody caseswere retested on the BPS withina 7-month span oftheoriginal testing.
None ofthe data changedsignificantly. In anotherstudy, six children in a non-adversarypopulationwere retested, and
the only change was for a child who was in family therapy,an expressed purpose of which to increase the quality of
parenting ofthe newcustodialparent. a
Validity hasbeentestedin several ways. jj One method involvedvalidationagainst a previouslyvalidated
instrument, the PerceptionofRelationshipsTest(PORT). Note: the validity methodofboth BricklinPerceptualScales
and the ASPECT test should be questionedasone methodof validitywhich appears to be highly touted bythe authors
ofthese tests is the fact that Judges pickedthe same person, a primary caretakerofthe child in a custodysuit, as the
person chosen by ASPECTor Bricldin,or the PORTtest. Knowingthe adversarialsystem as well as we all do, much
dependsupon the quality oflawyeringand the waythe case was handledin situationswhere custody is determinedby
theCouxt The author ofthis paper holdscertainreservationsabout this particularmethod ofvalidity. However,ifthe
validity data includedseveral hundred cases, thoseconcernscould be eliminated.
The second method ofvalidity measure utilized in Bricklin's PerceptualScales involved two kinds of child
questionnaires in which each child was asked to name theparent more likely to lend assistance,support,or controlin a
wide varietyof circumstances,and one whichrequired the child to describe what each parent would do in prescribed
practicaland fantasysituations. These last two methods had a 70% and 87%validity,respectively. Bricklin Perceptual
Scales,at 43.
Bricklin acknowledgesthat some experts havenot been gathering verbal responsesdue to their reasoningthat
forcing the childto give a verbal responsemaycause tension ifthe nonverbalresponseisdifferent. Bricklinassertsthat
BPS responses are rarely "fake"because "BPSnonverbalresponsesallow childrento tell what theywant to tellwithout
ever havine to come into conflict with parentalverbal commands." BPS Manual SupplementNo.4 10/15/86,
Bricklin admonishesevaluatorswho encountera childwhoroboticallypunches one end ofthecontinuumorthe
other to instruct the childas follows: "We alreadyknow you want to live with (whoever). This test is designedto give
us information. Spread yourresponsesout and give honest answers." BPS Manual SupplementNo. 4 10/15/86
Someinterestingvalidity research has been done in the short time that the Bricklinhas been out.
out ofthirty-sixcases in whichthetest asgiven went to Courtto determinecustody. In twenty-sevenoftheTwenty-nine
twenty-nine
cases, the judges, using all evidencepresented,including life histories, vocationaldata, school and medical records,
testimony offriendsandrelatives,etc., gavecustodyto the sameparent as the BPSscoredas beingtheprimarycaretaker,
(Id.at33-4.)
In the supplement,Bricklinnotes that in formal thejudge's choice and the BPS resultsofwho would
be thebetter primarycaretakingparent had an agreementhearings,
rate of89 percent. The same agreementrate was arrived at by
comparisonof results on other psychologicaltests and the BPS. The agreementrate between the psychologists's
interpretations and the clinical life history data was 91 percent. The agreementrate between the primary caretaking
parentand all available tests was 97percent. The surveywas basedon 27 respondents,usinga samplepool of141 cases.
BPS ManualSupplementNo.52/15/88.
6. Advantagesofthe BPS
The Bricklin PerceptualScales are so new that it is noteasy to assessitsadvantagessince the research
data on this test is so limited, unlike, for example,the MMPI or the Rorschach. However,it is clearfollow-tip
that the Bricklin is
much more of a valid tool for the clinician who is faced with evaluating one or both parents and making a
recommendation to the Court. Other personalitytests may reveal that Dad is an alcoholic, for example,but it does not
32
revealhow this trait inDad impacts on his child, or ifitis even relevant. Additionally, what has an impact on one child
does not necessarilyhaveimpact on another.
Dr. DavidI3rodzinsky, Associate ProfessorofClinicaland Developmental Psychologyat Rutgers University,
states the followingabout the BPS:
Despitethe limitations ofthese alternativeassessmentprocedures[such as the BPS], they representa valuable
addition to the field ofchildcustody evaluation. Most important, they shift the focus from a more traditional
clinical assessment to one in which the evaluator is focusingmore on a fUnctional analysis of the parties'
competencieswithinspecificchildcare roles. As such, these instruments are likelyto provide informationthat
is particularlyrelevant tothe issues beforethe court. Futureresearch,however,willneed toresolvethedilemma
regarding validation ofthese measures.
David M. Brodinsky, "On the Use and Misuse of Psychological Testing in Child Custody Evaluations,"
PROFESSIONALPSYCHOLOGY: RESEARCH ANDPRACTICE,vol. 24, no.2, 1993. p.214 -219.
As theBPS andother testsbecome more frequentlyused,theirreliabilityandvalidityincustodyevaluationswill
becomemore clear. To date,the Bricklin PerceptualScalesand the PerceptionofRelationshipsTest(PORT) havebeen
administeredmore than50,000timesaspart ofacustodyevaluation.They havebeenused inall fiftystates and accepted
as part ofcourtroomproceedings in these states. They have beenpurchasedby every level ofgovernmentfrom county
to the UnitedStatesgovernment. Moreover,hundreds ofmajor hospitalsanduniversitiesuse the BPS and PORTintheir
outpatientclinics.
Dr. Bricklin believes that the BPS successfullyaddresses all the main complexities which plague custody
decision-making.(Jjat 16.) At the very least,it is a flfl at addressingwhat is really relevantin custodydisputes: the
child.
7. Cross-Examination ofthe Expert
Anycross-examination ofthe expert shouldelicit the information that the BPS is a very new testingdevice and
has not yet received outside confirmation ofreliabilityand validity.
The attorneyshould pay attention to theage ofthe child that the BPS is given to, because even Dr. Bricklin
suggeststhat the 'bottomline"age, to insureforreliableand validuse, tobe age six and above. Ifthe child is eight years
ofage or below, the attorney should ask the psychologistifdevelopmentallythechild is capable ofabstract thinking.
Additionally, thetest may not bevery reliable forchildrenover the age of12, becauseifthat childis campaigningtobe
with a particularparent, the questionsare structuredin such a waythat a bright 12 year-oldwould realize the result of
his or her answers and, therefore, answer in such a way to make that particular"win." Ask the psychologist ifit is
possible that this has happened.
Additionally,ifvideotapingtheadministrationofthe testisnot possible,thena detailedexaminationofwhether
the clinician could, in any way, have influencedthe child's answers should be undertaken. Depending on how the
questions are asked the clinician, in administering the test, may greatly influence the child'sanswers.
B. Sex Abuse Leeitimacv Scales (Sals)

I. Parental AlienationSyndrome
The Sex AbuseLegitimacy Scaleswere developedbyRichardA.Gardner,M.D. Dr. Gardner,a practicingchild

psychiatristand adult psychoanalyst, is a professorat Columbia Universityin New York. Dr. Gardnerhas published
extensivelyon the subjectofpsychotherapywith childrenofdivorce. In his book, The Parental AlienationSyndrome
and the Differentiation between Fabricated and Genuine Sex Abuse, Dr. Gardner discusses the developmentand
underlying theory supporting his Sex AbuseLegitimacyScales(hereinafterreferredto as "SALS"). [R. Gardner,The
Parental Alienation Syndrome and the DifferentiationBetween Fabricated and Genuine Sex Abuse, (New Jersey:
CreativeTherapeutics,1987).]
Dr.Gardnerdevelopedthe SALSto focuson and battle what he hasdubbed the "parentalalienationsyndrome."
(jj at 67.) aDr.disturbance
Gardnerdefines "parentalalienationsyndrome"as:
in whichchildren are
preoccupied with deprecationand criticismofaparent - denigrationthat
isunjustifiedand/orexaggerated. Thenotion that such childrenaremerely 'brainwashed' is narrow. The term
brainwashingimplies that one parent is systematicallyand consciouslyprogrammingthe child to denigratethe
other. Theconceptofthe parentalalienationsyndromeincludesthe brainwashing component,but ismuch more
inclusive. It includes not only consciousbut subconsciousand unconsciousfactors within the programming
parent that contribute to the child'salienation from the other. Furthermore, (and this is extremelyimportant),
it includesfactorsthat arisewithin the child - independent ofthe parental contributions- thatplay a role in the
33
and the environment that may play a rote in bringing about the disorder." (
development ofthe syndrome. Inaddition,situational factors maycontribute,i.e. factors that existin the family
at 67-S.)
In laymen's terms, the parental alienation syndrome is when a child, for no valid reason, is distantand/or has
dislike for one ofhisor her parents.
Dr. Gardner contendsthat there are two reasons for the increase in manifestationsof the parental alienation
syndrome:(I)the"bestinterestofthechild" presumption superseding thesexistandoutdated"tenderyears"presumption
(i.e. the latter representing the theorythat mothersshould be favoredover fathersas custodiansof children oftender
years), and (2) withjointmanagingconservatorshipstaking an upswing,parents have been more apt to "brainwash' a
child to insurevictory, withoutconsideringthat the child mayin factbe betteroff with the denigratedparent. (jj
at 68-
9.)
Dr. Gardner notes that parental alienation will manifest itselfwith the "brainwashed" child an
"obsession"about the "hated" parent. Given the proper forum i.e., the court-appointedpsychologist'sdeveloping
office,judge's
chambers, orin open court,thechildwill provide"a command performance," givingadissertationon the alleged wrongs
committedby the "hated" parent in aspeech that has a rehearsedquality. The child'salienation may be rationalized by
trivial events such as the "hated" parent forcing the child to write out an incorrectspellingword 25 times to insurethat
the same error will not be repeated,or speaking loudly when telling the child to brush his or her teeth before bed, or
telling the child to get somethingfor the hated parent. Gardner states that:
"The professions ofhatred are most intense when the children and the loved parent are in the presenceof the
alienatedone. However,whenthechild is alonewith the allegedly hatedparent,he or she may exhibit
from hatred, to neutrality,to expressionsoflove and affection. Whenthese childrenare with the hated anything
parent,
they may let theirguard down and start to enjoy themselves. Then, almostas ifthey have realizedthat they are
doing something 'wrong,' they will suddenly stiffen up and resume their expressions of withdrawal and
animosity. Anothermaneuver commonlyseen in this situationis the child's professingaffectionto one parent
and askingtheparent to swear that he orshe will not revealto the other parent the professionsoflove. And the
provokingbecauseofthe fear that theirmanipulationsand maneuvers willbe divulged." (

same statementismade to the other parent. In this way these children 'cover their tracks' and thereby avoid the
disclosure of their schemes. Such children may find family interviewswith therapists extremely anxiety
at 71-2.)
Dr. Gardnernotesthat theproximityoftheloved parenthas an impacton thechild's reactions,and the closer the
loved parent is (in the room, outside in the waiting area, at home), the more likely the child will manifesthis or her
alienation. Dr. Gardner urges that anyoneinvolved in custody litigation recognizethe phenomenon. (i
at 72.)
2. FactorsContrihutinE to Development ofParental Alienation.
Dr.Gardneridentifiesthe following four factors as contributingto the child'sdevelopmentofparental alienation:
(a) Brainwashing
Brainwashing includesconsciousacts of programming the child againstthe other parent. Forexample:

(i) Dad convincing himself that Mom is having an affair, and telling the child about
innocentconversationsMom'shad with men as "proof' ofMom's infidelity.
(ii) Momcomplainingtothechild or children somuch aboutnewfinancialrestrictions that
thechildrenthink they may go without food,clothing or shelter becauseDad's now a
Scrooge.
(b)
(iii)
(
Talking to the children about problemswith the other spouse and exaggeratingthem,
such as: ifMomhas a glassofwineover a businesslunch,she becomes"an alcoholic."
at 75-88.)
Subtle Programminn
Subtleprogrammingis akin to brainwashing,butoften unconscious, An exampleofsubtle programmingwould
be the old guilt trip trick. Example: sayingto the child,"your fatherhas got other people in his life, all I have in the
wholeworld is you. Whenyou're gone, I am so alone". (j at 80-8.)
(c) Factors Arising WithinThe Child
The parental alienation often developsfrom contributionsarisingfrom "psychopathologicalfactors withinthe
child." For example: the child believing for some reason that he or she is "bad for wanting to visit" the noncustodial
parent or a daughterbelievingthat her father must choose betweenher and his newgirlfriend. (jj
at 89-92.)
34
(d) Situational Factors
Situational factorswould bethose". . .external eventsthatcontribute' and/or" .abetthe internal psychological

processesin the parentsand in the child." (Jj at 92.) Examples ofsituationalfactorsthat add to the parentalalienation
syndrome would be when one sibling observes another being punished, treatedbadly, orrejected for speakingout on
behalf ofthe rejectedparent. (IL at 92-6,)
3. Fabricated and Bona Fide Sex-abuseAlleeations in Custody Disputes

Dr. Gardner discussesthe changes ofthe late '70's and early 80's,ofcourts shiftingaway from the tenderyears
presumption towardjointcustody,whereby "an attemptwas madeto providea more egalitarianrole for both parents in
their children'supbringing." (jj. at 99.) Dr. Gardnercontinuesthat perhapsbecause of"a progressiveerosion of the
mother's secure position in custody disputes . . .[in 1982 or 1983 1 began seeing a new development,namely, the
utilization offabricatedallegationsofchild sex abusein the contextofcustodydisputes." (jj
at 100.)
Dr. Gardneracknowledgesthat children have been exposed to sex abuse as an increasinglycommon topic on
television, andthat sex abusepreventionprograms havebecomestandard in more schools,beginningevenatthe nursery
schoollevel. Both parentsand childrenare educatedand havethe ideathat an allegationofsex abuse will certainlyget
the Court's attention in a custody suit. As Dr. Daniel C. Schumandiscussedat an annual conferenceofthe American
AcademyofPsychiatryand Law in 1984, "[h]eightened instinctual forces in childrenand regressive loosening ofpre-
litigation characterdefensesin adults,both in the contextofstressful family breakdown,combineto generategenuine
perceptionsof abuse but invalid reports." (D. Shuman, "False Accusationsof Physical and Sexual Abuse," Annual
Conference ofAmericanAcademyofPsychiatryand the Law, Nassau, The Bahamas,October26, 1984.)
a. Criteria for Constructionofthe SALS

Dr. Gardnerpropoundsthe following criteriaforassessingthe child,the father, and the mother:
1. THE CHILD
Dr. Gardneridentifiesten criteriaas being very valuablein differentiatingbetweengenuineand fabricated sex
abusewhen interviewing the child.
(i) Presenceof Parental AlienationSyndrome.
Fabricating children are thoughttobe more likely to manifestsymptomsofparental alienation

syndrome.
(ii) Receptivityto Divulgence.

Kidswho are genuinetendnot tobethrilledbytheprospectofspillingtheirguts tocaseworkers,
psychologists, lawyers,judges,etc. Theoppositeis truewith fabricators. Fabricatorswill talk, andareoften encouraged
to do so by theaccuser,to anyone whowill listen.
(iii) Providing Specific Details.
Gardnerstipulates thatchildren whohavegenuinelybeenabusedwillpresentmoreconcrete facts
than children who are fabricating.
(iv) Credibilityofthe Description.

Ifa child is makingthingsup, it shows in his or herdescription.".. . [1]t is in the description of
theejaculate,especially, that the fabricator is likely to provide preposterousexplanations." (Gardner, !pfl at Ill.)
Moreover, the fabricating child is likely to take the interviewer's lead, and describe the "stuff" that came out of
perpetrator's penis as yellow and clear like urine --the only other "stuff'with which the kid has a frame ofreference.
(v) Guilt Relatingto the Consequences ofthe Disclosureofthe Accused.

Fabricatorsdonot tendto exhibitguiltor remorseover what mighthappen to theperpetratoras
a resultofthedisclosure. Childrenwhohavetrulybeenabused"may feel guiltyover theirdisloyaltyand therecognition
that the disclosure is goingto result in formidable painfulconsequences for the perpetrator.
(vi) Guilt Relatingto Participation in SexualActivities.

The fabricatordoes not feel guilt over the sexualactivitieshe orshe has allegedlytakenpart in.
The childwho has trulybeen abused may feel guiltyas a result of feeling pleasure over said activities.
35
(vii) Fear ofthe AllegedPerpetrator.

The fabricator usuallyis not frightened ofthe alleged perpetrator -- only ofwhat the accused
mightdo to the child as a result ofthe false accusation.
(viii) Sexual Excitation.
Children whohavetruly beenabusedexperiencean early "turnon" forwhich theymay seek an
outlet, such as physical contact with the interviewer.
(ix) Desensitizing Play.
Children whohave genuinely beenabused mayengageindesensitizing play --such as ejecting
the doll that representsthe perpetratorfrom the dollhouse.
(x) Attitude Toward Genitalia.
Victims ofabusetendtoconsidertheirgenitalsashavingbeendamaged. Fabricatorsdonothave

such feelings,and will not independently characterizetheir genitalsas damaged. (j4 at 109-17.)
Ofmoderatesignificance to Gardnerare thefollowing:

(i) the litany offabricatorhas a rehearsedquality;
(ii) genuinevictimsfrequentlymanifestdepression;
(iii) genuinevictimstend to withdrawfrom involvement with others;
(iv) sexuallyabusedchildrenareoften compliant, developingacheerful facade toward offthe threat
ofthedire consequences outlined by the perpetratorfor noncompliance;
(v) fabricators borrowtheirscenarios from other experiences,such as classroom presentations,
movies, etcetera;
(vi) victims tendto be more likely to suffer from psychosomaticdisorders;
(vii) children who have been abused tend to display regressivebehavior; and
(viii) children whohave been abused often displaya deep-seatedsense ofbetrayal. (j at 117-21.)
Dr. Gardner recognizes that sleep disturbances,chronicity, pseudo-maturity,seductive behavior with the
perpetratorand retractionmay all potentiallyhave high value in differentiatinggenuinefrom fabricatedsex abuse.
at 121-24.)
(jj
2. THE MOTHER
(i) Initial Scenario

The intervieweris to determinehow mama first learned of the abuse -- did the child report that dad
attackedhimin the shower, or did Mom, for example,inquire whether Dad had washedhis penis and ifhis penis "got
a little hard when Daddywas washing it?"
(ii) Shame
Gardner notes that many mothers would be ashamed at having to report reprehensibleconduct on the
part of their husbands, ifin fact the husbandhas truly abused the child.
(iii) SeekingHired Gun Evaluator
Gardner thinks that mothers of fabricatorswill, in the words of one of Dallas' top family
litigators,"hire her a whore.' Rather than trusting the court-appointedpsychologist,the mother ofthe fabricatorwill
bring in a hired gun.
(iv) Joint InterviewCorroboration.
36
Gardner has noted that fabricators, during the course ofjoint interviews with Mom, will
frequentlygive Mom side-longglancesto "check"their stories.
Gardnerpoints out the following criteria as having a moderate affect on differentiating mothers of children
fabricating sex abuse as opposed to mothersofchildren who have been genuinelyabused:
(i) Themotherofa genuine victim will appreciate the psychologicaltrauma sufferedby the child
as a result ofrepeated interrogations;
(ii) Mothersofchildren whohavebeenabused maystillrecognizethe importanceofthefather-child

relationship,and seek to preserve it in spite ofthe abuse;
(iii) Gardner argues that mothersoffabricatorswere not typicallyabused as children, butwill take
intoaccountthe childhoodhistory ofsex abuse ofthe parties;
(iv)
Gardner,mothers offabricatorstend to be aggressive and outspoken. (
While mothersofchildren genuinelyabused tend to be passive or incapacitated,accordingto
at 124-31.)
be assessed. (
Gardner observesthat the personalitycharacteristics regardingthemotherofthe child alleging abuse should also
at 131-32.)
3. THEFATHER
While Gardneradmits that generalizations regardingthepersonalitycharacteristics of fatherswho arebona tide
abusers are dangerous, he outlines the followingas very valuable differentiating criteria to distinguish a father of a
fabricator from a father who has abused his child:
(i) Bribes or Threats.
Children who have truly beenabused have frequentlybeenbribed or threatened by the erring father.
(ii) Indignation.
Fathers who have been falsely accused suffer fromextremeindignation.
(iii) Presence OfOther SexualDeviations.

Gardner argues that a man who engages in pedophilia will exhibit other deviations such as
'exhibitionism rape, voyeurism,sado-masochism, and homosexuality." Here Gardnerdisplays one ofhis biases: he
considers whathe characterizes as "obligatory homosexuality' as deviant. (jlj
at 132-34.)
Gardner pointsoutthe followingcriteriaas havingamoderate affect ondifferentiating fathersofchildren
who have beengenuinelyabused:
(i) Fatherswho have a history ofhaving beenabused themselves are more likely to abuse their
children;
(ii) Fathers offabricatorsare morelikely to beenthusiasticabouttaking a lie detectorthan someone

who has abused their child;
(iii) Fatherswho abuse are more likely to have a history ofdrug and/or alcohol abuse;
(iv) Fatherswho abuse are more likely to suffer fromlow self-esteem than fathers who do not;
(v) A father who has abused his child will be morelikely to regress in stressfulsituations;
(vi) Anabusermay choose a careerthatbrings himinto contactwith children; and
(vii) Fatherswho abuse their children are morelikely to be social isolates. (lj at 134-37.)
Gardner observesthat stepfathers are morelikely toabuse than natural fathers. He furthernotesthat fathers who
abuse, often rigid and strict, tend to be very moralistic about sex. (1c at 137-38.)
37
b. ClinicalEvaluation
Dr. Gardneradvocates the use ofa clinical evaluationin conjunction with the SALS. Dr. Gardner
use ofaudiocassettesand videocassettes to document theevaluation,and neveragrees tonot employsthe
being allowedto interview
the child in the presence ofthe allegedperpetrator. In fact,he mandates a court order allowinghimto interview all of
the parties, in any sequence he feels is warranted. Priorto interviewing a particularly youig child, he suggests that the
interviewer inquire oftheparentspet names that may be used for body parts. Gardnersuggests a direct verbal inquiry
with a child, giving the child an opportunity to draw a pictureand tell a story about it, andasking the child to drawa
picture ofa person, anotherofa person ofthe opposite sex, a third ofa family, and finally,he requiresthe child to tell
credible. (
a storyabout the familypicture. Unlessthe evidence or clinical examinationwarrants it, Gardnerstates thatthe use of
anatomicallycorrect dolls is simply "loading the dice," making information elicited through the use ofsuch dolls less
at AddendumHi)
Dr. Gardneralso suggests thatthe court-appointed psychologistresearch the medicaland/orhospital file on the
child, as by the time the evaluatoris called in, physical manifestations may no longerbe present.
c. The SALS Test
Many psychologistsfeel that Gardner's Sex Abuse Legitimacy Scale is not a "test." The reason: lack of
empiricaldata. Dr. Gardnercomes to his conclusions based solely on his clinicalexperience as opposed to anystudies.
Therefore,the SALS is best characterized as aclinical tool. Moreover,the SALS includesthe following warning on the
frontofthe instrument:
"WARNING: In order to be used in a meaningfulway, this instrumentmust be used in association with the
information provided by Dr. Richard A. Gardner in chapters 3,4, and 5 of his hook, The Parental Alienation
Syndromeand the Differentiation Between Fabricated and GenuineChild Sec Abuse (Cresskill,New
CreativeTherapeutics,1987). The book explains how bestto evaluate and scoreeach ofthe itemsin theJersey:
scale.
Failure to use these guidelinesmay result in misleadingor erroneous conclusions." (jj,
at Addendum III, 3.)
In the instructionsto the SALS, Gardneroutlines the followingsalientpoints:
.When medical evidence is not present, the SALS Scores may be the primary sourcesofinformation about
whetherthe sex abuse allegationis valid." (Iclj
1. ITEMS
The SALS consists of 26 items to be propounded to the child who alleges the sex abuse; II
items to be
propounded to the accuser, especiallywhere the accuser is the mother;and 13 itemsis tobe propounded tothe accused.
Theitemsand criteria are outlined above, but as an example, one very important item regarding the child
alleging the
sex abuse would be the item "very hesitantto divulge the sexual abuse," to whichthe interviewer would cheek "yes,
"no," or "not clear or not applicable."
2. CRITERIA
As outlined above, Dr. Gardnerhas created three categories:
(a) Very Valuable DifferentiatingCriteria
(b) Moderately ValuableDifferentiating Criteria
(c Differentiating Criteria ofLow But PotentiallyI-TitherValue
3, SCORING
Dr. Gardnerexplainsthe use andscoringofthe itemsas follows:
"The itemsare wordedsothat thegreaterthe numberofYesanswers,thegreaterthe likelihood that thesexabuse
is genuine. In contrast, thesmaller the numberofYes answers, the greaterthe likelihood the sex abuse has been
fabricated. Thedifferentiatingcriteria are divided intothreecategories, fromthe most to the lease valuable. In
order to give greater weight to the more valuable criteria, the followingpoint scores are to be given for Yes
answers in each ofthe threecategories:
Part A. Very Valuable Differentiating Criteria - 3 points foreach Yes answer.
Part B. Moderately valuable Differentiating Criteria - 2 points for each Yes answer.
I
PartC. Differentiating Criteria ofLow But Potentially Higher Value - point for each Yes answer."
(ith)
Separate scores are calculatedfor the child, who can score a maximum of 60 points, for the accuser, who can
score a maximumof27 points, and for the accused, who can also score a maximum of27 points. A cumulativescore
is not computed.
38
Psychological Testing ChapterS
In computingthe scores, scoresareweighted as follows forthe child,the accuserand the accused, respectively,
with "n' representingthe number of"yes" checks,as follows:
Child
Part A: n x 3 = maximumof39
Part B:nx2=maximumofl6
Part C: n x I = maximumof5
Maximumtotal ofA+B+C = 60
6 or below indicatesfabrication; 7 to 29 are inconclusive; 30 suggestsreal abuse
Accuser
Part A: ox 3 = maximumof 18
Part B: nx2maximumof8
Part C: n x I = maximum of I
Maximumtotal ofA+B+C =27
3 or below indicatesfabrication; 4 to 13 areinconclusive; 14 suggests real abuse
Accused
PartA: n x 3 = maximumof 12
PartB: n x 2 = maximumof 12
PartC: n x I = maximumof3
Maximum total ofA+13+C = 27
3 or below indicatesfabrication;4 to 13 areinconclusive;14 suggests real abuse

Dr. Gardner contends,based on his years of experience and research, that a score of 50% or more out of the
maximumor more suggestsbona fide sex abuse, whereas a score of 10% ofthe maximumorbelowsuggests fabricated
sex abuse. Ifone ofthe inconclusive score leans towarda significantscore, that increases the likelihood offabrication
or abuse,dependingon which end ofthe spectrum the score leans toward. ()
THE SALS SHOULD NOT BE USED AS A QUESTIONNAIRE. "Rather,the scale shouldbe used alter the
interviewswith the child, accuserand accused have been completed. Both individual and joint interviewsmust be
conductedin order toproperlyassess conflicting data that is often presented." (jj)
d. Reliability/Validityof the SALS
IftheSALSis a reliabletest, then any two psychologistswhoadministerit shouldgetthesameresults. There
areno studiesthat havebeen conductedregardingthereliabilityoftheSALS, however,consideringthesubjectivityof
thescoring oftheSALS, it would not be surprisingto finddifferentresultsdependingon theprejudicesand bias ofthe
individualpsychologist.
As a rating system,or a clinical tool, the Sex AbuseLegitimacyScales succeedas a valid instrument. In fact,
itcertainlycomescloser to measuring theexistenceofsex abuse than any othertestbeforeit. Ifforno other reasonDr.
£ Gardnershouldbeapplaudedforhis efforts. However,as ameansofcompilingdata capableofstatistical interpretation,
theSALSfails. Until Dr. Gardnerchoosesto publishthe empiricaldatahe hasgathered,andon whichhebases theitems
and scoringofthe SALS, thereisno wayto determineifthe SALS trulymeasureswhethersex abusehas occurredornot.
In fact,Dr. Gardner himselfnow testifies that it is not a test and has withdrawn it as such.
e. Direct Examination OfThe Expert
(1) Establish that the SALS is a systematic effort to quantify a very emotional situation in an objective
manner.
(2) The SALS takes into account the suspicion that sex abuse may havebeen raised merely for litigation
rather than for legitimate purposes.
(3) Establish that the SALShelps to determineif the parties and/or the child requirecounseling,and/or if
visitationshould be supervisedor limited.
39
PsychologicalTesting Chapter5
(4) Establish that the SALS providesa means for, and advocates, interviewingall of the parties involved
in thecase/sex abuse allegations.
(5) Establish that the SALSprovidessomemeansofhelp forthepsychologistindeterminingwhetherachild
has been sexually abused.
(6) Establish that the SALS strivesto befair in including the alleged perpetrator,thevictim andthe accuser.
(7) Establish that the SALS includes amultiplicityof factors under each area, allowing the interviewerto
check unclear or uncertain"(wluch is often the way things are).
(8) Establish that the SALS makes an effort at quantification by attributing numerical values to the
assessments, Beforethe SALS, no other psychologicaltest has providedsuch assessment.
f. Cross-Examination OfThe Expert

(I) TheSALS is nota true test. It is simplya ratingscale. Ifthe expert testifiesthat the SALSis a test, and
not simply a clinical tool, point out the lack ofempirical data validatingthe items or the criteria by which they are
adjudged.
(2) There are no standards defined for making the scale other than those outlined in chapters 3-5 of
Gardner's book. Ifthe psychologistdoes not have the same amountofexperience as Gardner, or ifhe or she did not
adhere to the guidelinesoutlined in the book, the resulting scores should be disregarded.
(3) Ifone of the parties declines to be interviewed then ask thepsychologisthow he or she can quantify
interviews ifone ofthe partiesdeclinedto participate?
(4) Question the psychologistabout Gardner's failure to show the building blocks through which he
established his items, or the justificationofhis '1,2,3' rating scale, other thanhis own experience.
(5) The SALS is subjective,and one examiner'sassessment could vary vastly from another examiner's
assessment. Additionally, Gardnerdoes not provide definitions forsuch wordsas "moralistic". What is moralisticto
one psychologistmay be blasphemy for another. Without such definitionsand standards to go by, the test can most
definitelybe effectedby the attitudes and beliefs ofthe examiner. The attorney should get the expert to admit to this
problemwith the SALS.
(6) Get the expert to admit that the rating scale forces dichotomouscategories of responseswhere most
responsesshouldfall on a continuum(cx: does the child lovehis or her parents?)
(7) Additionally,get the expert to admit that Gardner'sitemsand scores are based on opinion. Opinions
are not equal to standards.
(8) Ask the expert ifGardner does not state thingsas universal truths that may in fact not be universally
acceptedtruths: cx: althoughGardnerdisclaimshaving sexistviews, he characterizesthe father as the "hatedparent,"
and the mother as the loved parent," based on a "reflection of' his "own observation.' (Also note his opinion that
"obligatory homosexuality" is an illness. Further, Gardnercharacterizes the abuserbywhether ornot he has pedophilic
characteristics, but never explains howhe definespedophilia.)
(9) The expert should be forced to admit that the characterization of certain items as very valuable,
moderatelyvaluable,or oflow but capable ofhigh importanceis neverjustified by empirical data.
4. Phallic Plethysmouraph
The PhallicPlethysmograph, affectionatelyalso knownas the "dipstick," "petermeter," and "hardo'meter,"
may wellbe the functional,modem-dayequivalentofthe Salem WaterTest (used to determine ifa person was a witch
orwarlockas alleged,iftheaccuseddrownedwhensubmergedin the"SalemWaterTest" theywere obviouslyinnocent;
ifthey floated, they were burned at thestake as only those guiltyas chargedcould survive the water test). ThePhallic
Plethysmograph, by contrastto the other tests discussedin this article, is aphysiologicalbehavioral assessmentdevice.
Thedevicemeasures:
changes in blood flow through the penis [to] detect sexual arousal. By showing photographsof sexual
stimulito subjects,it is possibleto determine their sexualpreferences. Child molesters, for example,mightbe
expectedtoexhibitsexual arousal atthe sight ofyoung children;homosexualsatthe portrayalofsomeone ofthe
40
samesex; and heterosexuals formembersofthe opposite sex. Itisthen possible topinpointfeaturesthat subjects
findmostattractive,such as hair and eyecolor, height,bust measurements, or age." (Roedinger, at 525.)
a. History And Background
The phallic plethysmograph was developedas an adjunct to biofeedback. Biofeedbackmeasures a person's
responses to stimuli and allows the person to learn to control his or her responses to those stimuli. Phallic
plethysmography has been developing over the courseof the last twenty years as a meansofmeasuringmale sexual
response.
Speakinggenerically,
"Plethysrnography may be definedas the measurement ofchanges in volume ofa portion of the body. Since
transientchanges in the volumeofmost partsofthe body are related to blood circulation, plethysmography can
determinechanges in bloodvolumein the part being examined." [K. Kedia,"Penile Plethysmography Useful
in DiagnosisofVasculogenicImpotence," Vol. 22, No. 3 Urology235 (September,1983).]
Phallicplethysmographyhas beenused to diagnoseand treat impotence. In addition,in the legal forum, the
Phallic Plethysmograph has been used to assess and treat sex offenders. Plethysmographycombinesthe use of an
electrocardiogram machine and some type of "strain gauge," or "constricting ring." [P. Malcolm, et. al., "Control of
Penile Tumescence: The Effects ofArousalLevel and StimulusContent,"Queen'sUniversity,Kingston Penitentiary
TreatmentCenter,Canada BehaviorResearch and Therapy Vol. 23(3) 273-280 (1985).]
b. Construction ofthe Test
Althoughseveral types ofmethodshavebeen conducted,typically,subjects undergothe following:
(I) The subject is seated on a recliner.
(2) The subject is isolated.
(3) Thetechnicianadministering the test iseitherin the room,orcapableofcommunicatingwiththe subject
by intercom. The technician might also be behind a two-way mirror to assure that the subject is
watchingthe slides presented.
(4) The penis might be wrappedin something like Handiwrapto assure cleanliness.
(5) A plethysmographiccuffis wrappedaroundthe base ofthe flaccidpenis. (Note: the cuffis akin to the
type ofcufftypically used to measure bloodpressure).
(6) The cuffis constricted(ifa pneumatic cuffis used, thetightening is accomplished by injectingair into
thecuff"producingan air pressureequal to meanarterialpressureplusone-thirdoftheblood pressure);
(7)
this procedureyields "reproducible" measurements. (Kedia, jjjj note 107,at 236.)
Every actionhas an equaland opposite reaction. An increase or decreasein the size of the penis has a
direct impact on the amountofair trappedwithin the pneumaticcuff.
(8) The subjectis shownslidesranging from clothedmales or femalesto nude subjects,to prepubescentor
blank slides. The cuffmeasuresthe erectile response ofthe subject. The slides may be accompanied
by an audiocassette, whichvary the severityofaggressionportrayedby theslide. (klj
Penile tumescence has been documentedas the best measure of sexual arousal in males. [M. Zuckerman,
"Physiological MeasuresofSexualArousalin the Human," InN.S. Greenfieldand R.A. Sternback(Eds)Handbookof
Psychophysiology 709-749(New York: Holt Rinehart& Winston 1972).] Hence, the Phallic Plethysmograph is an
attemptto documentpenile tumescence. (Malcolm,jI2.a)
The above-described method is only one ofseveral available;actual plethysmographymeasuresthe volume
displacedby the enlargedpenis. (14.3 Somemethodsemploy circumferentialmeasuringdevices,includingmercury-in-
rubberstraingauges, [J. Bancroft,H. Jones & B. Pullman,"A SimpleTransducerforMeasuringPenile Erection,With
Comments on Its Use in the Treatment of Sexual Disorders", 4 Behavior Research and Therapy, 239-241 (1966).]
volumetric devices, [D.WheelerandH.B.Rubin, "AComparisonofVolunietricand Circumferential MeasuresofPenile
Erection", 16 ArchivesofSexual Behavior, 289-299(1987).]and metal-bandstrain gauges. [D.H. Barlow, R. Becker,
H. Leitenberg & W.S.Agras "A Mechanical StrainGaugeforRecordingPenile CircumferenceChange,"36 Joumalof
AppliedBehavioralAnalysis,73-76 (1970).] There has not been a sufficientamount ofcomparativedata compiledto
determine if one method is more reliable than another. The gauges all measure increased gauge strain, directly
corresponding to circumferential increases of the penis. Note: if ever there was an area calling for a case by case
analysis,this is it. Obviously,the transformation from flaccid to erect will be vastly differentbetweensubjects. Many
researchers therefore advocate conversion of circumferentialgauge scores to percent full erection scores, thereby
eliminating individual variances.[K. Freund, "A Laboratory MethodofDiagnosingPredominanceofHomo-andHetero-
Erotic Interest in the Male", I BehaviorResearch and Therapy, 85-93 (1963).]
41
Psychological't'esting Chapters
c. Measuring ErectileResponses
Sciencehas surelyprogressed since Mae West first asked"Isthat a bananain yourpocket,orare
to see me?" The followingmethods areused in measuring responsesto the Phallic Plethysmograph: you just happy
(1) Millimeter change in circumferencefrom the flaccidstate.
(2) Conversion ofraw score into percentagescore basedon subject'smeasurementat full
erection.
(3) Z-scoresare determinedby analyzingeach subject's responsestoall stimuli; mean and
standard scoresare calculated,thenconvertedinto "z" scores,whichrepresentthemean
in standardized deviationunits. [V.L. Quinsey and G. Harris "Comparison of Two
Methods of Scoring the Penile CircumferenceResponse: Magnitude and Area", 7
Behavior Therapy,702-04(1976).]
Researchersconflictoverwhen toscorethe erectileresponse--many for tumescencein terms

ofpeak response,while othersmeasurethetotal curve oftheresponse,whichopt scoringpenile
alsomeasureslatency. (Freund, jjji.)
d. Use Of Microcomputers In TumescenceMonitoring
Many innovatorsofPlethysmography advocatethe use of"computerassistedtumescence monitoringsystems"

[W.R. Farrall and R.O. Card, "Advancements in PhysiologicalEvaluationofAssessment and Treatmentofthe Sexual
Aggressor", 528 Annals of the New York Academy of Science, 266 (1988).] The new affordability of personal
computersmeans that every practitionercan have advancedtechnologyat his or her fingertips. "Preliminaryfindings
indicatethat with improvedprocedures, a wealth ofpreviouslyunseen data about the offendercan be gathered. (j4)
(1) Examples OfDeviantAudio Segments

Audio segmentsare used in the presentationofsome Phallic Plethysmographs, followedby, or in conjunction
with, slides"involvinga wide varietyofdeviantsexualactivitiesfocusedon male/female victimsover a wide range, and
an adult-heterosexualsequence." (hjat 268.) Whiletheauthors refuseto presentexamples ofdeviant visual segments
used in a Plethysmograph, following are two examplesofdeviantaudio segmentsformulated by Robert D. Card, ofthe
Clinic for Counseling and Psychotherapy, Inc., and William R. Farrall, of Farrall Instruments,Inc. (note: Farrall
Instruments is the world's largestmanufacturer ofplethysmographs),who have developeda new,computer-controlled
stimulus presentation,data collection and analysissystem:
CHILD MOLEST, FemaleVictim3-8 years Old. I really like littlekids. They like me too. They always seem
to want to climhon my knee or play horsie with me. I feel so good when I see them running tome when I go
into the room. Theywant to climbon my lap and wrigglearound and cuddleup to me. They'reso soft and cute.
I hopeshe wantsto sit on my lap today. She's thecutestone. I feel so goodwhen I'm there. Ijust want to touch
her a bit. Maybe we can do it in the other room. It feels so good whenshe squirmsaround. Maybe she'll want
to touch it. I could gether to do some thingsifwe playedsome games. She seemsto know all about itsoit can't
hurt her.
ADULTHETEROSEXUAL, Adult female. I'm really having a hard time concentratingon my work today,
She'swearingthat tight sweateragain. Ifshe comes over here once more and bends over my deskI'm not sure
I can keep from grabbingthem. I can almost feel theexcitementif! could pushmy facebetweenthem. Get a
hold ofyourselfi She looksalmost as good from the back. Can't keep my eyes offher butt as she walks
past.
I can almost feel my hands runningalong that soft body. No wonderI'm so homy! I haven't had it for days.
Thelast time we got it on was wonderful! I've got to get some tonight. I can almost feel her squirming with
excitementat the touch ofmy hands. (jj)
e. TreatmentUses
For those who have been convicted of rape, child molestation and/or a related offense, the Phallic
Plethysmograph canbe usedto treatand lessendeviantresponsesbyjuxtaposingarousingpictureswith aversiontherapy
--suchaswhiffsofammonia.[R,E,Freeman-Longo and R.V.Wall, "Changinga LifetimeofSexual
Crime",Psychology
Today, 58-64 (March, 1986).]
One practitionerin Dallas who administersthe Phallic Plethysmographfor federal probationand state child
protectionagenciescommentsthat the PhallicPlethysmograph is". . . notthe be-all,end-all,butit'sthebest tool we have
now for evaluatingthese cases." [S. Crawford, "Device Used for Offenders", The Dallas Momintz News, 41-42A
(November 13, 1988).] Child protection agencies in both Dallas and Tarrant County have used the Phallic
Plethysmograph"when trying to determine whether to allow a father accusedofchild molestingback into the home."
(Id. at 42A.)
42
WilliamFarrall,president ofFarrallInstruments,Inc.,the world's largestmanufacturerofplethysmographs,has
been quoted as saying:
We keepthem as faraway from law enforcementas possible. Thequalityofjustice is less than perfect, and we
don't want this used to force confessionsout ofpeople. ()
While Mr. Fan-all may have the best of intentions,deviant responses to the Phallic Plethysmographshave
undoubtedlybeen used by prosecutorsin even our fair-minded county to push an accused perpetrator over the brink
towarda full confession.
Some researchershave noted the limitationsand properuses of the Phallic Plethysmograph:
• . First the penile tumescence plethysmographis nota sexual 'lie detector.' It will not tell whethera suspected
offenderhas actuallyoffended,nor will it tell ifthe offender is certain to offend again or not to offend. In other
words, it cannotbe used to searchforprobableoffendersin the general public. 1-lowever,properly obtainedand
interpreted, the penile evaluation can generally make it possible to determine the gender preference, age
preference, and in many cases the type ofsexual activityand interest to both an offender and a non-offender.
This is obtainedby noting the relativelevel ofsexual response tovariouslevels ofstimulus materials." (Farrell
& Card, note 120, at 262.)
Of extreme importancein assessing results of Phallic Plethysmograph is the realization that "sophisticated
offenderscancontroltheirerectionsandoftenalter theirresponseduring assessment. Less knowledgeableoffendersmay
team from the assessment experience or are coachedby those whohave previouslybeenassessed." (jj
at 264.)
f. Reliability/Validity
The physiological studies done from the 1960's to the present regardingthe Phallic Plethysmographmay give
thedevice an auraofcredibilitythat it does not deserve. Almostall studiesregardingthePlethysmographconcludethat
there isa need for furtherresearch. Althoughthe Plethysmograph has some value as a treatmentdevice, questions arise
as to the validity ofthe test outside ofthecontext ofthe laboratory.
"The key questionto the validity ofpenile erectionmeasurements in diagnosingsexual deviants remainsthe
comparison ofthose findings with the response ofa normal populationto the same erotic stimuli. Only then
couldtheexaminerassertwithreasonablecertaintythat the subjectwho deniedhavingaparaphiliacdisorderbut
showedapositivearousalpatterswas in fact tellingthe truth." [S. Travin, K. CullenandJ.T. Melella,"The Use
and Abuse of Erection Measurements: A ForensicPerspective," Vol. 16, No. 3, Bulletin of the American
Academyof Psychiatryand the Law, 235,241(1988).]
InstrivingtowardvalidationofPhallicPlethysmography,mostresearchershavecomparedandcontrastedresults
of"deviant" subjects, such as the results ofchild molesters,to the results ofother deviant subjects, such as rapists:
"Unfortunately, there have been few studies that have comparedlarge normal subjectgroups with parapiliacs.
The majority of comparative studies have tested intragroup differences, e.g., aggressive pedophiles versus
nonaggressive pedophiles . . . there were no significant differences between the erection measurements or
incarcerated rapists and incarceratednon-rapists." (fflj
Currentresearchcertainlydoesnot supportthe theorythat resultsofPhallic Plethysmographsshouldbe used as

a meansofprovingguilt or innocence ofa sex crime,or that a fathershoutd be deprived ofvisitation rightsbecauseof
deviant scores regardingpedophilic slides. If used as part of a comprehensiveanalysis of a given subject, Phallic
Plethysmography may havesomemodicumofcredibility. Perhapsthemostappropriateuse ofPhallic Plethysmography
is in aversion therapy. However,whenused in the context ofallegations ofsexual molestationin a familylaw matter,
the reliabilityand validityofthe instrumentshould be seriouslyquestioned.
g. Direct ExaminationOf The Expert
(I) Havetheexpert establishthe subject'ssexualhistory,andpreviouspsychologicaland criminalrecords,

ifany.
(2) Havethe expert establishwho supervisedthe PhallicPlethysmograph, and theconditionsunderwhich
thetestwas administered.
(3) Have the expert explain how measurementswere taken - is the reportbasedon raw scoresor
percentages/z-scores?
(4) Havetheexpert report whetherthe accusedever reacheda full erection and ifso, in responsetowhat?
43
(5) Have yourexpert explainwhether the penile tumescence scores were measuredby a pneumaticcuff,
a volumetric device, a mercury-strain gauge,etc. and whether the experts opinion ofreactionswould be any different
ifany oftheother means had been used.
(6) Havetheexpertreportthetimesequencebetween audiocassette deviantpresentationsanddeviantslides,
ifboth were used.
(7) Have the expertreport ifan audio presentation was used, whetherexplicitdetails offoreplayand were
intercoursegiven, and ifthe detailswere violent.
(8) In thecaseofan accusedrapist,havethe expertdifferentiatereactionsofa rapist from anon-rapist?from
someonewho engages in pedophilia? from a homosexual? from a non-offender? How?
(9) Also your expert should explain how he orshe would recognizefaking.
Ii. Cross Examination OfThe Expert
(I) A subject may score a wide variety of falsenegatives and false positives on the Plethysmography-
doesn't the range ofpossible inaccuraciesmakethe test totallyunreliable?
(2) Inquire as to whether the test taker was subjected to a blood exam prior to administration of the
Plethysmographs? (Drugs and alcoholmay haveadirect impact on results. Ifa father in a custodydisputewas accused
ofchild molestation,and took a Valiumprior to submittingto thetest, the results could be skewed).
(3) Inquire as to whether the recordingequipmentused was standardized?
(4) Ask the expert: ifanother expert with his or her credentialswas shown these test results, is there any
guarantee that he or she would interpretthe subject'sresponses in the same fashion you have?
(5) Inquire as to whetherthe expert would characterize visual materials as more stimulatingthan audio
materials? (CardandFarrall would assertthat audio materials are more stimulatingsexuallythan visual materials,due
to theimaginationfactor).
(6) Ask the expert ifit is true that sophisticated offenders can controltheirerections? Aren'tsophisticated
offenderscapable ofaltering their responsesduringassessments?
(7) Ask the expert ifthere is any wayto know ifthe test taker is thinkingabout the slidepresentlyin front
ofthemwhenan erectionoccurs,orwhetherthetesttakermaybe daydreaming about apreviousslide, or whethera prior
slide has causedthe subject to lapse into a fantasy.
(8) Inquire as to whether the subject was allowed to become detumescent between audio-visual
presentations. Ifnot, can the scores the subjectreceivedas the Plethysmograph progressedbe consideredaccurate?
(9) Ask the expert ifhe or she considers the Plethysmograph as accurateas a lie detector? (Ifnot, the
Plethysmograph resultshaveno more businessbeing admittedthan lie detector results).
(ID) The expertshould be questioned as to whetherthe testresults be consideredindependentofa thorough
reviewofthe subject'ssexual history. If so, it is time to impeach the expert!
(II) Question theexpert asto whetherhe orshebelievesthat thePlethysmographresultsshouldbe the sole
basis for decidingwhetheror not the subjectshouldbe released from prison or from a treatmentprogram? (Ifyes, you
have an expertready to be impeached).
5. Custody quotient (Cci)

a. Historyand Background
The CustodyQuotient (hereinafterreferredto as the "CQ") is a new approach to assist psychologists in the
evaluationofparents in childcustodycases. The CQ was developed by Dr. RobertGordonandDr.Leon Peek,ofDallas,
Texas, as a guide forexplainingexpert findings by report and incourt. Further,thedevelopersoftheCQsee it as aguide
for planning theremediationofa parent's shortcomings and how to use theparents strengthsto the best of his or her
ability. The authors ofthe CQ designed it becauseofa growingdissatisfactionwith traditionalpsychologicaltests as
they are appliedto measuring good parenting.
b. Purpose
Theprimarypurposeofthe CQisto assistCourts,attorneys and impartialthird parties to resolvechild custody
disputesin a child'sbestinterest. The CQ is alsodesignedto haveresults that are relevantto determiningsole andjoint
custody,includingtheassigningofparentingrightsand dutiesas wellas decidingissuesofaccess. Additionally, theCQ
results address the threshold issue ofwhethera material change ofcircumstancehas occurred in a parents life whena
modification has beenfiled.
Anotherpurpose oftheCQ istoaddto psychologist's fundofassessmentproceduresa device whichisespecially

designedto addresscustody questions. The developers ofthe CQ felt that while there are a number ofvalidand reliable
psychologicalproceduresfor assessingpsychopathology,few havebeen researchedfor legal applicationand fewerstill
for custody issues.
44
c, Construction oftheCQ
The CQ is a systemofmapping judgments(similarto DSM HI-Itjudgments)or thejudgments of a diagnostic
team about a person's capacityto bean effective parent. The CQ is not intendedto directthe decisionsofthe court or
jury regarding the custodyofa child or children,but to assist in providingrelevant information about theknowledge,
attributes, and skills of adults as well as provide information relevant to the issue of modifying existing custody
arrangements.
Additionally,the CQ leads to designingremedial programs for parentswith low scores and therefore the CQ
scoreand profile isnotstaticbydesign. The parentmayretaketheCQ anddocument progressmadeonestablishedgoals
every six months. Unlikemostpsychologicalinstruments,the person takingthe test has an opportunityto study, learn,
gain experienceand then improvetheir score.
The CQ's Manualoutlinesan education-based therapy for acquiringnewparenting skills and for retraining a
parent with impoverished skills. In the absence of remediationor a significantevent in the life of the parent, a CQ
evaluationisconsideredcurrent, by its authors,for up to six months. Followingremedial therapy,retestingat intervals
of three months is recommended. (It. Gordon and L. Peek, "The CustodyQuotient: A Test ofEffectiveParenting to
Assist with CustodyDecisions',TRIAL INSTITUTE,The Texas AcademyofFamilyLaw Specialists, 1989.)
Gordon and Peek did something that to the authors' knowledge has never been done. Gordon and Peek
researchedTexas lawsbeforeallowing any item on the CQ. Therefore,there is legal authorityfortherelevanceforeach
CQitem from case law, statuteor by reasonableinferencefrom the UniformMarriage and DivorceAct.
Each parent response is graded with referenceto written standards (similar to the Wechsler Comprehension
Scale). In doing so, the examiner may considerother information acquiredabout the parent and child. The Frankness
Scale on the CQ helps the examinerdecide the degree ofcandorwith whichthe parent disclosedinformation (similar to
the L, K and F scales on the MMPI). The maximum time allowedfor administering thestandardCQ interview is two
hours. (j43
Presently,the classification ofCQ results are derivedfrom the theoreticalpropertiesofthe normal curve. For
that reason,Gordonand Peekwarnthat untilnationalnormsare available,inferencesabout goodparentingbased on CQ
results aloneshould be made with appropriatecaution. (jj)
The CQisa set ofratings basedon a compositeofclinicalprocedures. The examinermay selectfrom a variety
ofstandardapproachespracticalfor theparticularcase: the standard CQ interview,a clinical interviewwith history,
objective and projectivetests, a reviewofdocuments,observationsofthe parent-childinteraction, a home study and
collateralcontacts.
The CQ avoidstechnical tenns and is designedtobe sex neutral. The commentsectionsin the CQManual points
out cultural and intellectual differences amongparties as they areknown to affect psychologicaltest conclusions and
thereforethe Manual pointsoutthat cultural differencesmust be kept in mind. (liij
Additionally,the attorney shouldkeep in mind that the summaryCQ scoredescribesthe parentata single point
in time. But itconsidersparent performance over-time. Aparent living in an intact family,who intendsto divorce,will
likely havea differentprofilesix monthsafterseparation.
Examiner qualifications for administering the CQ is the American Psychological Association guidelinesfor test
users. The CQmay alsobe administered by a trained assistantunder qualifiedsupervision. (kjj
d. DefmitionofGood Parenting
The CQ Manual offers the followingdefinitionofgood parenting:
"Thosecollections ofattributes,skills andbehaviorswhich adultsrely onand use in raisingthe nextgeneration.
Good parenting occurs when adult practices lead the child to live independentlyand fulfill their biological,
psychologicaland social potential.
The followingpsychological laws ofgood parenting are offered,
Law I. Goodparenting is designedto protectthe child fromharm duringthe child'svulnerableperiodofgrowth
and maturation.
Law2. Goodparenting includesteaching thechildskills formasteryover their environment so the child canlive
independently of theirparentsand after theirparents have died.
45
Psychologica'Testing Chapter5
Law 3. Good parentingcreates an environment for the child conduciveto the child fulfillingthe
biological,
psychologicaland social potential.
Corollary 1, Thoseattributes,skills and behaviorsofa parent which facilitate the operation ofthese laws are
in the best interest ofthe child.
Corollary2. Thoseattributes,skills and behaviorsof a parent which interfere with the operation ofthese laws
are not in the bestinterestofthe child. fR. Gordon and L. Peek,
In CQ tenninology,the adjectives "good", 'effective' and "competent' are used interchangeably.
e. DevelopmentOfThe CQ Scales
The CQ consistsoften clinicalscales and one additional scale, The developmentofeach scale was done in the
followingfashion:
(I) ExpertOpinion
Theexpert opinionofthe developersoftheCQ was used to definea number ofelementsin the domainof
good
parenting. (BothGordon and Peek are licensedpsychologistswith considerableexperience in examiningparents and
children in custody disputes. In addition,Dr. Gordon is trainedin law and
ProfessorPeek has taught psychometric research at the graduatelevel for many years).
(2) Theory ofChild Psychology

Additional elementsofthe domain ofparenting were gleaned from a review of theory and research in child
psychology,childpsychiatry and childdevelopment. Thereviewalso includedliteraturein the fields ofanthropology,
philosophy,theologyand comparativepsychology. (For details seeThe CustodyQuotient- Research - Manual,Chapter
2). Q3
(3) Home Studies

A systematic observationofseveral characteristicsofthe observablejobofparenting was conductedbased on
video-taped home studies.
(4) Opinions ofParentsand Children

Several studies were done ofattitudes ofparentsand childrentoward goodparenting and related issues. (For
details, see The CustodyQuotient - Research - Manual, Chapter2).
(5) Opinions ofJudges and Attorneys

A survey was conductedof districtjudges and family law specialists (that survey is in the process of being
updatedand presently approximately 450 district judges have respondednationwide).
(6) Based in Law
Gordon and Peek eliminated any item of parenting for which there was no authority in law to indicate its
specificallySec. 12 of fl
relevancyfor custody decisions.The family codeofvarious states,case law, the UniformMarriageand DivorceActand
Texas Family were reviewed. (See Chapter2 ofQ Manual for more detail).
(7) Categories
The refmed itemsofgood parentingwerebinnedinto categories ofparentingknowledge,attitudes,abilities, skills
and behavior. A descriptive word or phrase was selectedto represent each scale ofitems,
f. The CQ Scales
Eachitem ofeach scalereceivesequalweight,however,due tothe studyconductedofjudges and
scales receivegreaterweight. attorneys,two
(I) EMOTIONALNEEDS SCALE(EN)

This scale consistsoften items that are directed toward how well the parent meets the emotionalneeds ofthe
child. An example ofthe EN Scale is as follows:
46
"EN 8 Parent willing to admitmistakes
2 parent has the maturity and insightto acknowledgeerrorsin parentingor in decision making about child
1 parent seems reticentto examine their own errors or shortcomings in parenting or lacks the insightto do so;
is overly defensiveor superficial
0 parentseemsincapableofrecognizingtheirownshortcomings orerrors;typicaldefensemechanisms are denial,
rationalization and projection (blameschild or the other parent or bad luck for thingsgone wrong)
What decisionsor actionsconcerningyour child are you most proud of!

Are there some decisions or actionsyou'vetaken you now think were mistakes?
Comment No oneis perfect. A matureparent is able to recognizetheir ownshortcomings and makedisclosures

about errors in judgment. Emotionalproblems in children are sometimescaused by parents insisting that all of the
failuresand disappointments ofthe childare due solely to shortcomings in the child. The too obviousexample is the
parent who suggeststo the child that the child is responsible for 'Daddyor Mommyleaving home."
For superior ratings,look to parents who have madea point ofbeing clear with the child that the divorce was
not the child'sresponsibilityand that thedecisionofpost divorce ormodificationplans is thejob ofthejudge and not
thechild.
It is helpful to ask the child iftheir parents say they're sorry when they make a mistake. It is not proper for
mental health professionalsto encouragechildrento "informt' on their parents." (jj)
(2) PHYSICALNEEDSOF THE CHILDNOW ANDINTHE FUTURESCALE

(PN)
This scale contains fourteenitems that were designedto assess how well the parent takes care ofthe physical
needs ofthe child. An example ofthe PN Scaleis as follows:
"PN 3 Provides healthydiet
2 parent reveals concernand awarenessofnutrition;providesbalanced diet; child occasionallyhelps prepare
meals
1 parent knowsand employsadequateconcepts ofnutritionand diet

0 parent usuallytakes child out to eat unhealthy fast food;orders out most nights;servesjtmkfoods
Who prepares your child's meals at home?

How often does yourchild eat out? How often do you take fast food home?
In general, what are themost importantaspectsofachild's diet? (Ifvague)Couldyou pleasebe more specific?
(For schoolage children)Does your childbuy ortake lunchto school? (Iftaken to school)Who preparesyour
child'slunch for school?
(Ifcooking is delegated)How did you pick the cook/housekeeper?How do you know they are doing a good
job?" (Ij)
(3) NOEMOTIONALORPHYSICALDANGERTO THE CHILDSCALE(ND)

The ND Scale like the ED Scaleis givengreater weightthanthe other scales. There are eleven itemson the ND
Scale, an example ofwhich follows:
47
"ND 8 Child'sSurroundings are Free From... Kidnapping/denying access
2 parentdoes not havepossession ofchild contrary to a court order; parenthas not frustratedor denied other
parent's access to child in person, by phone or mail contrary to court order
I parent expresseswillingness to kidnap child or to frustrate otherparent's access to child contrary to a court
order
0 parenthas unlawful possession ofchild; parenthas denied lawful access ofotherparentto child
(For custodialparent) Are theresomecircumstances under which youdo not allowyour child to see or talk to
theirotherparent?
Howdo you keep the otherparentinformed ofevents, problems, programsand activities in your child's life?
(For non custodial parent) How often do yousee your child and talk with your child by phone?
Is there a court orderconcerningtheperiodsoftime your child is in your home? (Ifyes)What doesitprovide?"
(4) GOOD PARENTING: KNOWLEDGEABLE, A1TRIBUTES, SKILLS,

ABILITIES,PARTICIPATION SCALE (GP)
This scaleconsists ofthirteenitemsand is designed to measure the parentingskills ofthe parent. An example
of the GP Scale follows:
"GP 9 Parent is able to organize family affairs
2 parent coordinates the schedules ofmembers offamily to complimenteach other; child getsplaces on time
I parent tries to organize competing routines of family members; there is occasional tardiness or missed
appointments; parent tends torigidly adhere to schedules so that spontaneous opportunitiesfor growth and learning are
missed
0 parentseems unable to toleratechallenge to familyorganization;familyevents are happenstance; child often

misses appointments or is tardy
What is your child's usual daily schedule in yourhome, during the week and on the weekend?
What is the most difficult part ofyour child's schedule to keep on time? What do you do when this problem
causes your child to run late?
CommentThe examinermay wish to ask whether the parentkeeps a calendar ofthe child's activities and those
of other members of the family. The examinermay want to review the familycalendar at a later session. Obviously
organizingchildren'sactivities is more criticalforschooldaysthan weekends or holidays. Conventionalwisdom is that
both chaos and overly compulsive organizationhave untoward results." (jj)
(5) PARENT ASSISTANCESCALE(PA)
This scale, consistingofseven items, is designed to determine how appropriatelya parent is using help from
others with their child. An example ofthe PA Scale follows:
"PA 2 Family support(grandparents/other siblings)

2 grandparents/relatives live close by, are available, and desire to help with child; siblings are healthy and
constructive
I grandparents/relatives are only tangentially involved in child's life due to distance,lack ofinterest or poor
health; are availableonly for celebrations ofmilestonesin the child's life or for holidays; siblings are good models for
child but are too busyto supportparent
48
0 grandparents/relatives are uninvolvedin child'slife and are preoccupied with theirown needs; grandparents
arcdeceased; siblings have impoverished relationship with child
Are yourparents living? (Ifyes) Wheredo they live?

When did they see yourchIld last?
Whatactivitiesdid they do with the child on that occasion?
Are thereother relativeswhosee your child fromtime to time? (Ifyes) Who are they? Flow often doall ofyou
gettogether?
Does your childhave a brotherorsister, or a stepbrotheror sister? What is the brotheror sister like? Flow do
they get along with your child? Give me an example ofhow the brother/sisterhelps you parent?
Give meanexample ofa way in which the brother/sisteris not the best influencefor yourchild?
Comment This item concernsthe parentsand relatives ofthe parentbeing interviewed and ofthe siblingsliving
with the parent. As the life spanofAmericans increaseand since halfthe American work force is comprisedofwomen,
grandparents and great grandparentsplay an increasinglyimportantrole in child'scare.
Manygrandparents havesubstantialcontactwith theirgrandchildrenand provide emotionalsecurityandasense
ofcontinuity. Although it is commonforbrothers and sisters and stepsiblingsto deny likingeach other orcaring about
one another,just the oppositeis ordinarily true.
In evaluating theinfluenceof an older sibling,commonissues are: maturity,power, knowledgeand loyalties,
privacy and self-demarcation. Ifthe older siblinghas replacedthe parent in terms ofresponsibilityforthe child,rate '0'
on this item. If thereare no older siblings then omit this item." (jj)
(6) PLANNINGFOR THE CHILD SCALE(P)

Consisting offive items, the P scale is designedto measure howwell the parent is planning for all ofthechild's
needs. An exampleofthe P Scale follows:
"P 3 Planningfor child's medicalneeds
2 parent knows nameofphysicianforroutine assistance; describesplan for helping child with special medical
needswhetherfordisabilityorallergies;describesprocessofselecting and evaluating specialistsandmethodofpayment
for futureservices
I Parentknows somethingabout child'smedicalhistorybut is weak on details; cannot give names or selection
processfor specialists
oParent does notknow physician;cannotrelate howhe/shewoldselectphysicianorgivespoorprocess(e.g.,use
whoeverfriendat work uses); seems obliviousto the importanceofroutinemedicalcare
Does yourchildhave any specialhealth problem? (Elyes) How are you goingto help themwith this problem?
Who does your child see to have annual physicals? How did you select the doctor?
(Ifnone) How would you go about selectinga specialistifone was needed?" (Ij)
(7) HOME STABILITYSCALE(HS)

As the name ofthe scale suggests,this scale is designedto measurethe stabilityofthe parent's home. The I-IS
scale consistsofsix items, an example ofwhich follows:
"I-IS 5 StableLifestyle
49
2 parentmaintainsconsistent employment, stable relationships with friends;parent maintainspredictable
scheduleor informs child ofchangesso childcan reach parent ifnecessary daily
I parent or situationhas forced at least one majorchangeduring the past five years; parent is not reachableby
child due to changingscheduleexceptat certaintimes ofday
U parent makes arbitrarychanges in pattern of living; parent frequently
changes circle of friendsin the home;
child does not know visitors; frequent changesdo not permitchild to know general activitiesofparent
What are the most important changesyou have made in your lifestyle since your divorce(or separation)? By
lifestyle I meansuch things asjobs, economic status, where you live, what you do for recreation, who you spend time
with.
Comment This item involves an overall impressionby the examinerofpreviousUS items. (j4j
(8) ACTS/OMISSIONS SCALE(NO)

The AJO Scale consists of eight items and is designed to measure parent misconduct. The following is an
example on the MO scale.
"Acts/Omissions
The personal standards ofthe examinerare not the guiding criteria for rating the parent on these items. The
criteria is the contemporarycommunity standard. Moreover, the examiner should refine rating on the community
standard in terms of any known subcultural or ethnic group differences. Also, the examiner must guard against
stereotypesbasedon gender whenrating the parent for these items,
Methodsofcorroborationinclude interviewing the childand the other parent.
NO I Parent misconduct
Misbehavior is any activityoutsideofthe limitsofwhat is accepted. This is differentand more permissivethan
what is avengeor typical to a group or community. Peoplewill tolerateactions more atypicalthan the behavior they
enact. Also, themisbehaviormust havea direct,inferredorpotential impact on the child inorder to be coveredby this
item.
2 no sign, indication, suspicion or report ofmisbehavior

1 misbehaviorhas occurredbut has been discontinued; parent taking steps to prevent repeat
0 ongoing misbehavior
Has anyonesuggestedthat you have done something improperor illegal duringthe past ten years?
(Ifyes) What did they say? Is what they said true or partiallytrue? What were the circumstances?
Was your child aware ofthe situation? (Ifso) Who told the child about it?" (j)
(9) VALUESSCALE(V)
This scale consists ofsix items and is designed to measure how well the parent is transmitting values which
allows the child to distinguishrightand wrong. The followingis an example ofthe V scale:
'Comment This item referstoan attitudeorbeliefthat the parent holds. Most valuesare transmittedtothe child
implicitlyorthroughrole modeling. Clearlythere are some mattersabout which a parentmust havethe last word such
as the dangerof traffic for young childrenor the use ofdrugs by teenagers,the schoolsubjects childrenshould take,or
thesports they areactive in.
50
V 2 Ethics ofparent
2 parent wordsand actions illustrate moral and ethicalvalues withoutcontradictoryactions
I parentonly occasionally voicesappropriatevaluesor behaves in a contradictory manner
0 parent shows or voices valueswhich contradictthose ofthe general community or denigrates the need for
values
Whatare the most important thingsthat you want your childtoknow aboutthe differencebetweenright
and wrong?
How do you teach these principles to yourchild?" (j4j
(10) JOINT CUSTODY SCALE(JC)

TheJC Scale is designedto measurewhether a parent is a candidateforjointmanaging conservatorship. The
scale consistsoften items. The following is an example ofthe JC scale.
"JC 8 Geographic proximity ofhomes
2 parent livesclose by otherparent;parent has realisticplansto providetravelbetweenthehomes (e.g., children
are old enough to walk or bike between the two homes in normalweather)
I parent voicesintentionto live near other parent; parent has weak plans for travelbetween thehomesby the
child
0 geographical relationship ofhomesdoes not allow for safe, convenientor inexpensivetravel
How far is it in minutesfrom your home to the other parent'shome?

Doyou or the other parenthave immediateplans to move?
Wouldyou be willingtomove ifit made it easier for yourchild to go from one home to the other? How about
moving out ofthis city if it became necessary?
Comment When parents live remote from one another or live in neighborhoods with highly disparate
socioeconomicstatus,joint custody arrangementsusuallyfail. Jointcustody arrangements are "usually"strainedwhen
one ofthe parentsremarries." (j43
(II) THE FRANKNESSSCALE(FS)

TheFS Scaleis not one ofthetenclinicalscales but insteadis an additionalscale designedto measurewhether
a parent has a proclivitymerely to answer thequestions in a socially desirablefashion. The scale is similar to the "L"
scale on the MMPI. When a parent gives too many answers to the frankness questions in a "socially acceptable"
direction,motivation distortionorperceptualinaccuracy is inferred. The FS questionsare dispersedthroughouteach of
the other scales. An example ofan item on the FS scale is as follows:
"GP 14fF Frankness
I Parent frankly admitsto occasions when he or she did poorlyin handlingchild's misbehavior
O Gives only superficialexamples or speaks in generalities
51
Are there timeswhenyou haveregretteddisciplining your childas you did? Forexample,because you learned
yourunderstanding ofa situationwas wrong oryourdiscipline was too harsh or too lenient?
Comment Appropriate disciplineis a matterofparental discretion, Usuallya parent has apreferredmethod for
infants - facial expression,raising voice or momentaryphysical restraint;for the young child - spanking,
withholding
privileges,"timeout", sendingthe childto their roomor the"Oneminutescolding";for teens - grounding,no telephone
or television, or restricting driving fora period oftime. The examinershould determinewhether the parent'spreferred
method is consistent, predictableand produces the desiredresult.
As a method ofcorroboration, it is helpful to ask the child whetherthey think their parent's is fair.
Thereis a difference betweena parentbeing authoritativeand being authoritarian. Effectivedisciplinediscipline
does not include
makingthe child think theyare a bad human being. Nor does it involveembarrassingthe child.
As is true ofother CQ items, the examinershould considerthe practicesofthe parent's subculturewhen
theparent on this item. rating
The authors define abusive discipline as parent reactionswhich cause physical injury to the child or which
diminishthe child'sselfesteem. For extremeparent response,refer back to NDI and ND2." (j4j
g. The Custody Quotient Score

The CQ summaryscore is expressed as a standard deviationquotientscore. This score is
relatively easy to
compareto otherpsychometric measures. Thesummaryscore lies onscale wheretheaverage forthe generalpopulation
is 90 to 110, "superior" beginsat 120, and "borderline"beginsat 79.
At this time, the "ClassificationSystem"is derived from the theoreticalproperties ofthe normal curve. The
normalcurve is the most common statistical distributionin nature, including establishedpsychologicalphenomena.
The property ofthe normal curve importantto the CQ scores is that a majorityofpersons are
withinplus orminusone standard deviation,closelyapproximatingthe averageofthe general population. expected to fall
A CQ score
of85 to 115 will contain about 68% ofthegeneral population. A CQscore of70 to 130 will contain about 96% ofthe
generalpopulation. (jtj, at 9.)
h. The CQ Classification System
CQIJQ Parent % ofParents

Grand Score Competency Range in Range
130 Very Superior 2.2

120-129 Superior 6.7
110-119 HighAverage 16.1
90-109 Average so.o
80-89 Low Average 16.1
70-79 Borderline 6.7
69 and below
() Dangerous 2.2
i. Validity And Reliability

It mustberememberedthat theCQispresentlystill in the researchstage and thereforethe studies ofvalidityare
presentlyunderway. The CQ should thereforebe treatedwith that in mind. However,unlike The Bricklin Perceptual
Scales (BPS) and The Sex Abuse Legitimacy Scales (SALS), studies are being conductedto determine
validity and
reliabilityand the CQ is consistentlybeing updatedas newdata comes in.
Specifications for the CQ call for a mixture ofcontent,constructand criterion validity.
52
(1) ContentValidity
Thedevelopersofthe CQ identified a domain ofknowledge, attributes,attitudes,skills and behaviors ofadults
which ate relevant to the tasks of measuringeffective parenting. The items themselvesgrew out of a review of
interdisciplinary researchandthroughthecollective experience oftheauthors. Theirrelevancewas corroborated through
the understandingsofchildren,parents,judges, attorneys and the CQ Panel on Development. (jj)
(2) ConstructValidity
Definitionsand laws ofgood parenting were offered to give specificityto the domain of good parenting. A
contentanalysisofthe understandings ofparents, children, attorneys and judges was madeto define the parametersand
boundariesofgood parenting the CQ instrument intended to address. Direct observation ofparents and children in a
professionaloffice environmentand through videotaped home studies was also relied upon. Constructvalidity makes
inferencespossible from CQ scores to the domainofgood parenting. (kU
(3) CriterionValidity
At this point it seemsunlikelythat winning a custody lawsuit will emerge as the generallyaccepted criterion
againstwhich to validatepsychologicaltests. There are no alternativemeasurementswhichcould be used as criteria.
(4) Classification
Theclassificationsystem for CQ ranges ofeffective parenting were derived from the mathematical principles
of the theoretical normal curve. The upper range was emphasized in order to enhance fairness during the CQ
developmentperiod. This approach parallels the construction of classification systems in natural science. When
normative studies are complete,the basis ofclassificationwill shift to empiricaldata. (jj)
(5) Item Format
theadministration ofthefirst 100 tests based on the commentsoftesttakers.

(6) ExternalValidity
(
Items andstandardinterview questionswere reviewed bythreepersons withknownwritingskills for appropriate
grammar, the eliminationofbiasin syntaxandconstructionflawssuch asambiguity. Questionswere refined following
at 10)
External validity was subjecttoa GeneralSystemTheory Approach(GSTA). The cross-examination technique

was employed. Sample CQ resultswere presentedand attackedin the mock courtroomofthe WilmingtonInstitute by
volunteeringmembersof the CQ task force. Basic rules ofevidencewere observed.
The General Systems Theory Approach postulates that traditional methods of establishing validity do not
considerthe Gestaltofan instrument that is intended to measurea complexsystem ofbehaviorsuch as goodparenting.
(7) Reliability
Thereliability oftheCQwillbeestablished byevaluating thedegree towhichtwo ormore expertexaminerswill
giveaparticularparent the sameCQscore andthedegreeofinternalconsistencythat existsamongCQ itemsand between
CQscales. A pilot study ofinterrater reliabilityfound high ratingreliabilitybetweentwo experiencedraters. (Ij)
(8) Research Status
The CQ is both a clinical instrument, which summarizes the ratings of expertsalong standard criteria, and a
psychometrictest. At this state ofdevelopment, it is a "clinical instrument'.
Clinical instruments serve as guides to experiencedexaminersand provide for a consistentway to record the
opinion of the experts. Clinical instruments are 'wide band' in that many attributes of the person are weighed and
assessed when the examinermakesthe ratings. A typicalpsychologicaltest by contrast, scores attributesin a more or
less mechanicalway. ()
(9) Standards and Guidelines
The CQisbeing developed with referencetothestandardsforpsychologicaltests oftheAmerican Psychological
Association. Application is being developedwith referencetoconstitutionprecepts ofdue process.
53
At this time the CQ has the specificityof administration typical of some well-dcsigned psychological tests.
Research is underway that will determine whetherthescoringofthe CQ is also sufficiently specific. This distinctionhas
to do with the type ofinformation assessed and thenature ofthe user. (klj
j. Direct Examination OfThe CQ Expert
(I) Establish with the expert that while the CQ is a research device, it represents probablythe best range
ofinformationthat shouldbe assessedin a custody battle.
(2) Have theexpertexplainwhetherthe resultsto the CQ represent new oroldbehavior for the subject,and
whether the primacyofthe behaviorshould have an impacton test results.
(3) The expertshould reporton whether he or she has based his or her opinionexclusivelyon the CQ, or
whether results to other tests and methods have also been addressedin reachingwhateverrecommendationis made to
the Court. (Ifthe CQ is given appropriately, many other factors aretaken into consideration by the expert).
(4) Establish the expert's knowledgeofthe ongoingvalidationprocessbeingassembledregarding the CQ.
(5) Establish that becausethe CQ isstandardized,the experthas guidelinetocomparetwo different parents.
(6) Establish that the CQ is a comprehensiveinterviewsystem.
(7) Itwould be helpfulto the fact finder to know some ofthe items on theCQ. Anyjudge orjury will see
therelevancyofthe itemson the CQ especiallyas comparedto "evil spiritspossess meat times' from theMMPI.
(8) Establish with the expert that no otherpsychologicalinstrumentofassessmenthasjustification for each
item in the law.
k. Cross-Examination OfThe CQ Expert
(I) When cross examiningthe expert who has given the CQ, and it did not come out favorable for your
client, establishthat theCQ is not a test.
(2) Theexpertshould be cross-examined about the fact that the CQ is notyetsupportedby normativedata.
(3) Get the expert to admitthat the CQ's rating system is presently unstipported by empirical data.
(4) Theexpertshould also admitthat at presenttheCQlacks external validity.
(5) Ifthe expert is using theCQ to say who should have custody,then get theexpert to admit that even
Gordon and Peek say that that isnot the purposeofthe CQ.
(6) Ifonly one parent took theCQ, there is no means ofcomparison, andthe expert should admitthat this
lessens the valueof the CQ.
5. Uniform Child Custody Evaluation System (UCCES)

Since no existingpsychological test has emergedas the test to be utilized in custody evaluations,and because
the existing tests are so dependent on the skill of the clinician in making reasonable interpretations in custody
determinations, Dr. Harry Munsingerand Dr. Kevin Karlsondevelopeda uniforminterview procedurefor all mental
healthprofessionals,calledtheUniformChild CustodyEvaluation System(UCCES). Initially, it isbeing publishedas
an interviewand decision-making guide while validationdata is collectedby thepublisher, PsychologicalAssessment
Resources. Additionally, theAmericanPsychologicalAssociation'sCommittee on ProfessionalPracticesand Standards
iscurrentlyin the processofdevelopingstandardsand guidelines for administering child custody evaluationtests such
as those previouslydiscussed.
The hope is that in a few years, the UCCESwillprove to be valid and reliablestandard assessment
for use in custody evaluations to that the primary focus ofcustody litigationcanonce again be the childrenprocedure
and their
parentsand not the tests and who gave them. This test should be availableby the end ofthis year.
VI. CONCLUSION
A. General Considerations When an Expert has PsychologicalTesting
Thefirstthing an attorney should be cognizantofis that, exceptwith rare exceptionlike the BricklinPerceptual
Scales, CQ, SALS, and UCCES,psychologicaltests were not designedto assesslegal issues. Psychologicaltests are
designedas a means to help the expert evaluate a patient, not to be taken into the courtroom and used as an end in
themselves.
Putting asidethevalidityofeachindividualpsychologicaltest,theattorneyshould evaluatewhethertheclinician

has used the test beyond its limits. Psychological tests, as stated several times in this article, are merely tools for the
clinician, and as such, their usefulnessis limited by the experience,expertise,and integrityofthe clinician and ofthe
54
If of
purpose ofthe test. the developer the testor the clinicianfail to understand the followingconcepts, then that failure
to understand can leadtoabuses ofthe tests and the test data:
I. The psychological tests which are used most in the United States are based on norms consisting of
middle class Caucasian individuals. Therefore, the test may be valid when given to middle class
Caucasians but invalid when given to persons outside ofthosecharacteristics. US!. at 62.)
2.
3.
(
All psychologicaltestscome with standard administrativeinstructions. The validity ofeach test is
dependent on there being no significant deviationfrom thoseinstructions. The eflècts ofthe deviation
ofthe test resultsmay be difficult to ascertain, at 62.)
How wellan individual performs on agiven test isoften determinedbythe setting in whichit is given.
Individualswith similarpsychologicalcharacteristics orintelligencemayhavequite different testresults,
dependent uponwhetherthetestis givenina noisywaiting roomofa clinician's officeor in a quiet room
offby itself. (j4. at 62.)
B. Standards and Guidelines for Child Custody Evaluations in Divorce Proceedings
Attorneysand psychologistsalikeshould be aware ofnationalchild custodyevaluationguidelinespromulgated

by the American PsychologicalAssociation which are attached as Appendix "A." The guidelines focus on the
psychologistsrole in the divorce proceedingand provide guidance and standards for the psychologistin preparingthe
child custodyevaluation. The guidelinesaddress three major areas:
I. The purposeofa child custodyevaluation;
2. Preparingfor a child custody evaluation;and,
3. Conducting a child custodyevaluation.
The major principle set out by the guidelinesreiterates the preceptfound in nearly every state's familycode or
statutes - the child's best interests and well-being are paramount. Applyingthis rule to child custody evaluations, the
guidelinesprovidethat theprimary purposeofa child custody evaluationis to assessthe bestpsychologicalinterestsof
the child. To bestachieve this goal, the focusoftheevaluationshouldbe on parentingcapacity, the psychologicaland
developmental needs ofthe child, and the resultingfit.
In preparingfor a child custody evaluation, the guidelines define the role of the psychologist as a objective,
impartial, professional expertwho does not attemptto act as ajudge. Moreover,the psychologistshould have current
and specialized training orknowledgein order to competentlyundertake a child custodyevaluation. The psychologist
should also be aware of any personal and/or social biases that he might have which could affect his opinions in the
evaluationprocess. Additionally, the psychologist should avoid any counselingor therapeuticroles with anyof the
partiesand confmehimselfsolely to his defmedtask.
In conductingachildcustodyevaluation,thepsychologistshouldlimitthe scope oftheevaluationtoissuesraised
by the referring person or court, and obtain informed consent from all appropriateparticipants. Informed consent
includesinformingthepartieipantsasto the limitsofconfidentiality anddisclosure oftenrequired incustodyproceedings.
Thepsychologistshouldalso use every relevant and helpful available resource, whileguardingagainstinappropriately
interpretingorassessingthe data. Finally,the psychologistshouldmaintaindetailedrecords inaccordwith the relevant
statutory guidelines.
While many ofthe individual guidelinesstate whatappears obviousto litigation-wise psychologists,theyserve

an importantpurposebecause they establishthe foundationand limits for a proper custodyevaluation. This provides
a reference point for the family law attorney who is overwhelmedand confused by the unfamiliar world of child
psychology. From the perspective of the psychologist,the guidelinesestablish coherent standards for divorce and
custody related practice.
C. FinalThoughts
It seems clearthatopinionsbased on psychologicaltests are,at best, controversialandsometimes may even be
unconstitutional. This does not mean that psychological tests are not effective tools for the clinician in a clinical
situation;however,it is time that the legal community questions theuse ofthe presently established psychologicaltests
in the courtroomand, mostspecifically, in family law courts. Thereare no studieswhichsuggestthat, ofthe standard
tests which are administered to custodyparticipants,there is any validityin using thoseteststo assess parentingskills.
Furthermore, there is an astoundingamount of evidenceshowing that the standardtests do not "meet the criteria of
established scientific principlesor general acceptance by the scientific community." (Ziskin, note 2, at 250.)
55
At the vety least, the attorney should use this article, and the sources cited herein, to becomeacquaintedwith
information beyond ournormal expertise. Withoutsuch information, an attorneycannot effectivelyperform director
cross-examination ofan expert who has administered any ofthese tests.
56

Psychological Testing

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Psychological Testing

Uploaded by

Copyright:

Available Formats

PSYCHOLOGICALTESTING

ROBERT GORDON, ID., PH.D., Addison

State Bar of Texas

DOCTOR ROBERT GORDON

Dr. Robert Gordon is a forensic and clinical psychologist and

Dr. Gordon is boarded in the specialties of clinical and forensic

The Wilmington Institute provides trial and settlement science

Wilmington Institute & Lw. Robert Gordon 2

MARYJOHANNAMCCURLEY: born Baton Rouge, Louisiana, October3, 1953; admitted

AUTHOR AND LECTURER:

Fellow - American Academy ofMatrimonial Lawyers

Author, PROTECTING YOURASSETS FROMA TEXASDIVORCE(2005).

DIVISION OF PROPERTYON DIVORCE,University ofHouston Law Foundation (April 2000)

U. GENERAL CONSIDERATIONS OF PSYCHOLOGICAL TESTING

IV. PERSONALITY TESTS 8

V. NEW DEVELOPMENTS IN PSYCHOLOGICAL TESTING WITHtNTHE LAST DECAl) 29

Psychological testingis a standardized methodofcheckingaportionofan individual'sbehavior and comparing

Reliabilityofa testisusuallydetermined in termsofa coefficient correlation betsveen test scoresona firsttesting

However,validity is more difficult to establish thanreliability. For example,assumefive witnessesalt testify

There are differentapproachesto validity, whichare:

ilL INTELLIGENCE TESTS

Anintelligencetestmay indicate a personality disturbancewhen thetest taker's performanceis erratic. Such

Intelligence tests cannever beusedalonetodiagnoseapatient,althoughthey cancontributeto theunderstanding

4. Observationsofthe Subject's Behavior

I. Historyand Construction oithe Stanford-I3inet Test

twelve, he or she would receive an I.Q. of 120.

C. Wechsler Adult Intelligence Scale — Revised or Third Edition (WAIS—Ri

TheWA!S-Rconsists ofelevensub-tests. Sixsub-tests are intheverbalcategory and fiveare in theperformance

(1) Similarities- Thelastsub-test ofthe verbalcategoryis thesimilaritiessub-test. The individual

Schafterbelieve that depressedindividuals tendto make low scores on this sub-test. (

(c) Vocabulary. The vocabularysub-test,although considered the best estimate ofintelligenceon

(a) Picture Completion. There isastrongculturalcomponent to the taskassociatedwith the picture

1. Minnesota Multiphasic Personality Inventory(Mmpi)

(1) "MMPI" - As indicatedabove,theinitials standfor Minnesota Personality

(2) "ITEM" - This is thetermused to referto the questions appearingintheMMP!.

(4) "1" SCORES AND"RAW" SCORES - A "raw" scoreon theMM?! indicates

(5) MALE/FEMALE SCALES - The designofthe MMPI recognizes that males

a. STEP I: SELECTING THE ITEMS: Hathaway and McKinley gatheredover 1,000

Additionally, according toreviewersinBurosFifth MentalMeasurementsYearbook,[O.K.Buros,Fifth Mental

Further, eventhoughthere is anewnormativesample, thesamplegroup of1,408 people are all formMinnesota.

According to BeverlyKaemmer,a publications representativeofthe UniversityofMinnesotaPress,notmuch

comprise randomsamples within communities chosenbecausetheirdemographic characteristics conformclosely

Drs. HathawayandMcKinley boiled downthe initialpool of1000itemsto 550 itemsorquestionsontheoriginal

(c) The Contemporary NormativeGroup: CoUigan, Osborne,SwensonAndOfford, 1983.

Colliganandhiscolleaguesdevelopeda newrange ofresponsesfor the MMPI. The Colligangroup alsocreated

6. Creating the MMPI-2 Scales.

Frequently scoredsupplementary scales on the MMP!-2 includethe following:

(c) MANIFEST ANXIETYSCALE(MAS).

(e) DOMINANCE (Do)

(i) OVER-CONTROLLED HOSTILITY (0-H).

Thisscaleattempts to distinguishassaultivefrom non-assaultive prisoners. Highscorershave strongemotional

8. Myths and Facts AboutScoring The MMPI And The MMPI-2.

9. MMPI-2 Profile Scoring

10. ComputerScoring ofthe MM!'! andMMP!-2

Additionally,16 ofthe MMPI itemsare repeats. Those itemsare:

R. B. Cattell, in anefforttoarriveata comprehensivedescription ofpersonality,began assemblingall personality

bestknown is the 16 PersonalityFactorQuestionnaire (Cattell,Eber, & Tatsuoka,1970).

Similar inventories havebeen developed for ages 12 to 18 (High School PersonalityQuestionnaire),8 to 12

The assessmentofresponsebias is especiallyimportantin forensicpsychologicalassessments and constitutes

a. Historyand Rackround ofthe Rorschach

"Reliability ofinterpretation is, ofcourse,themost important matterofall. Itmay be stated as a general

Therefore,BarryRitzier,Kevin SharkeyandJamesChudyputtogethera newsetofpictures;however,they used

individual personality traitsaccurately.

(1) Rotter Incomplete Sentences Blank Test (Risb)