Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 1

PSYCHOLOGICAL ASSESSMENT | CHAPTER 5 – VALIDITY

DEFINING VALIDITY
VALIDITY – defined as the agreement between a test score or measure EVALUATING VALIDITY COEFFICIENTS
and the quality it is believed to measure. 1. LOOK FOR CHANGES IN THE CAUSE OF RELATIONSHIPS
 Test measures what it is supposed to measure.  conditions of a validity study are never exactly reproduced.
 1985 – AERA, APA, NCME published a booklet entitled  The logic of criterion validation presumes that the causes of the
Standards for Educational and Psychological Testing. relationship between the test and the criterion will still exist
 Standards were revised in 1999 and 2014 when the test is in use. However, there may be circumstances
 Latest version of standards is organized in three sections: under which the relationship changes.
foundations, operations, and applications. o Example: A test might be used and shown to be valid for
o Foundations is similar to our principles and focuses on selecting supervisor in the industry; however, the validity study
basic psychometric concepts such as validity and may have been done at a time when all the employees were men,
reliability. making the test valid for selection supervisors for male
o Operations considers how tests are designed and built and employees. If the company hires female employees, then the
how they are administered, score and reported. test may no longer valid for selecting supervisors because it may
o Applications takes on a wide range of issues, ranging not consider the abilities necessary to supervise a sexually
from training required to administer and interpret tests. mixed group o employees.
 Validity is the evidence for inference made about a test score.
2. WHAT DOES THE CRITERION MEAN?
ASPECTS OF VALIDITY CRITERION-RELATED VALIDITY
1. FACE VALIDITY – is the mere appearance that a measure has  Studies mean nothing at all unless the criterion is valid &
validity. reliable
 A test has face validity if the items seem to be reasonably  Constructors attempt to correlate their tests with other tests that
related to the perceived purpose of the test. have unknown validity.
 Example: a scale to measure anxiety might include such as  Meaningless group of items that correlates well with another
“My stomach gets upset when I think about taking tests” and meaningless group remains meaningless
“my heart starts pounding fast whenever I think about all of the
things I need to get done”. Can we conclude that the person is REVIEW THE SUBJECT POPULATION IN THE VALIDITY
anxious? In this case, we can only conclude that the person STUDY
answers these two items in a particular way.  Validity study might have been done on a population that does
 Face validity is really not validity at all because it does not not represent the group to which inferences will be made
offer evidence to support conclusions drawn from test scores.  Example: Some researchers have debated whether validity
 in many settings, it is crucial to have a test that “looks like” it is coefficients for intelligence and personnel tests that are based
valid. These appearances can help motivate test takers because primarily on white samples are accurate when used to test
they can see that the test is relevant. African American students
 Example: Industrial settings, attrition can seriously jeopardize
2. CONTENT-RELATED EVIDENCE FOR VALIDITY – considers the validity studies.
adequacy of representation of the conceptual domain the test is designed  Those who do poorly on the job either drop out or get fired and
to cover. thus cannot be studied when it comes time to do the job
 Example: if you are being tested on the first six chapter of this assessment.
book, then the content to the test is related to the information in  If there was a group that did well on the test but failed on the
the chapters. job, then it might not be represented and could be systematically
 Content validity evidence offers some unique features, such as it eliminated from the study because the workers were already off
is the only type of evidence besides face validity that is logical the job by the time the assessment came around.
rather than statistical.
 Two new concepts relevant to content validity evidence: 3. BE SURE THE SAMPLE SIZE WAS ADEQUATE
1. Construct underrepresentation – describes the  Problem to look for is a validity coefficient that is based on a
failure to capture important components of a small number of cases.
construct.  Sometimes a proper validity study cannot be done because
2. Construct-irrelevant variance – occurs when scores there are too few people to study
are influenced by factors irrelevant to the construct.  Cannot depend on a correlation obtained from a small sample,
particularly for multiple correlation and multiple regression
3. CRITERION-RELATED EVIDENCE FOR VALIDITY – tells us
 Such a study can be quite misleading.
how well a test corresponds with a particular criterion.
o CROSS VALIDATION STUDY - good validity study will
 A criterion is the standard against which the test is compared.
present some evidence for this; assesses how well the test
 The reason for gathering criterion validity evidence is that the actually forecasts performance for an independent group of
test or measure is to serve as a “stand-in” for the measure we are subjects
really interested in.
 INITIAL VALIDITY STUDY: assesses the relationship
between the test and the criterion
TYPES OF EVIDENCE FOR CRITERION VALIDITY ARE:
 CROSS VALIDATION STUDY: checks how well this
1. PREDICTIVE VALIDITY EVIDENCE – forecasting function of tests.
relationship holds for an independent group of subjects.
 Example: the SAT Critical Reading Test serves as predictive
validity evidence as a college admissions test if it accurately
forecasts how well high-school students will do in their college
studies.
o Predictor variable – quantitative and verbal tests
o Criterion – college GPA

2. CONCURRENT VALIDITY EVIDENCE – comes from assessments


of the simultaneous relationship between the test and the criterion.
 Example: learning disability and school performance.
 VALIDITY COEFFICIENT – relationship between a test and
a criterion is usually expressed as a correlation. This coefficient
tells the extent to which the test is valid for making statements
about the criterion.

You might also like