Professional Documents
Culture Documents
Instrumentasi PembentuknKesahanKebolehpercayaan
Instrumentasi PembentuknKesahanKebolehpercayaan
Instrumentasi PembentuknKesahanKebolehpercayaan
Pembentukan Instrumen,
Kesahan dan
Kebolehpercayaan
Instrumen
Experiments
Social surveys
Cross-sectional
Comparative (cross-national)
Longitudinal
Content Analysis
Secondary Statistical Analysis
Official Statistics
Demography
Epidemiology
Field stimulations
Structured Interviews and Observation
Powerpoint Templates
QUESTIONNAIRE PLANNING
1. Determine the Objective
What do you want to ask and of whom?
Consider how the data will be analyzed. What information is
needed?
2. Determine the Mode of Administration
Paper and pencil
Cost effective, but low response rate
Web / E-mail based
Cost effective, higher response rate, some suspicion of
the medium
Combination
Can be administratively cumbersome
Powerpoint Templates
QUESTIONNAIRE PLANNING
Powerpoint Templates
CONSTRUCTING the
QUESTIONNAIRE
Consider what objective each question measures
Three to five questions per objective is recommended
How many questions?
Find a balance: long enough to obtain the necessary
information but short enough so respondents wont lose
interest.
Questionnaires that are too long wont be completed
Goal: 25-30 minutes MAX
Powerpoint Templates
DESIGNING THE
QUESTIONNAIRE
Question Formats: Open Ended
Questions
Allow respondents to express feelings and expand on ideas.
Question should be phrased to avoid a one word response.
What aspects of the Sociology program best prepared
you for your current position?
Open ended questions at the beginning of the questionnaire
may be discouraging to respondent include one or two at
the end
Analysis of responses can be challenging
Powerpoint Templates
QUESTIONNAIRE
DEVELOPMENT
Question Formats: Closed Questions
Ranking
Forces the respondent to rank order responses based on a
value judgment.
Rank the following items with regard to how you spend time at
Bucknell from 1 (spend the most time) to 5 (spend the least
time).
_____ Studying
_____ Socializing
_____ Volunteering
_____ Working at a part time job
_____ Exercising
Powerpoint Templates
QUESTIONNAIRE
DEVELOPMENT
Question Formats: Closed Questions
Checklist
A number of possible answers are provided and the
respondent is asked to choose one or choose all that apply.
How did you finance your Bucknell University education?
Choose all that apply.
_____ Parental contribution
_____ Student Loan
_____ Work Study
_____ Off campus employment
_____ Other (please specify): _____________
Powerpoint Templates
QUESTIONNAIRE
DEVELOPMENT
Question Formats: Closed Questions
Scaled items
Very common
Indicate strength of agreement or disagreement with a
statement.
Numbers are assigned to each response on the continuum
The English curriculum at Bucknell University adequately
prepared me for my current position.
1 y
2 e 3 / No 4 5 gly
ngl gre t ral on ree o n
ro
St gree Di s a u
Ne pin i Ag Str ree
a O Ag
Dis
Powerpoint Templates
QUESTIONNAIRE
DEVELOPMENT
Question Formats: Closed Questions
Suggestions for writing scaled items
Use positively worded or negatively worded items
consistently
Use an odd number of responses along the continuum to
allow for a neutral response
Anchors may be different based on focus of the item
(Strongly Agree Strongly Disagree; Frequently Never;
Very Good Very Poor)
Keep items short
Powerpoint Templates
QUESTIONNAIRE
DEVELOPMENT
Question Formats: Categorical
Typical demographic questions:
Yes/No/?, Male/Female, F/S/J/S/G
Answers fall into certain categories respondents pick the one
that applies to them
Analyzed using frequency, proportion, percentages
2 considerations:
Questions on a continuum should stay on a continuum if they
will be directly analyzed
Example: Age allow respondent to write in their age
rather than check off a box for their age range
Categories must be collectively exhaustive & mutually
exclusive (next slide)
Powerpoint Templates
QUESTIONNAIRE
DEVELOPMENT
Problematic: Corrected:
Number of hours spent Number of hours spent
studying per week: studying per week:
18 21 18 21
21 24 22 25
24 27 26 29
Categories are not
mutually exclusive
Powerpoint Templates
IMPORTANT
CONSIDERATIONS
APPEARANCE AND DESIGN
Have clear directions that are age appropriate & easy to
follow
Format, size, reproduction should facilitate completion
No less than 12 pt
Times, Arial, Courier
PILOT STUDY
Give the questionnaire to a few people to try it out
Have them evaluate readability and clarity of questions
Estimate time required to complete the questionnaire
Suggestions for revisions
Powerpoint Templates
SENDING IT OUT
FOLLOW UP PROCEDURES
Send to all participants to preserve confidentiality (If you
havent already returned the questionnaire..)
May include a second copy if available
Usually done in 2 week intervals
RESPONSE RATE FOR MAILED QUESTIONNAIRES
Typically low - 30% is considered good
Higher response rate with web-based questionnaires
Good follow-up increases response rate
Powerpoint Templates
SUMMARY OF STEPS
1. Determine the Objective (consider analyses)
2. Determine the Mode of Administration
3. Determine the Sampling Methodology
4. Construct the Questionnaire
5. Institutional Approval
6. Conduct the Pilot Study
7. Write the Initial Communication
8. Send the Questionnaire
9. Follow up
10. Analyze the Results
Powerpoint Templates
DEVELOPING QUALITATIVE
INSTRUMENT
Powerpoint Templates
Examples of Qualitative Research
In-depth Interviews
Focus Groups
Ethnography/Field Research
Historical-Comparative Research
Discourse Analysis
Narrative Analysis
Media Analysis
Powerpoint Templates
Reliability and Validity
Consistency of Measurement
Reproducibility over time
Consistency between different
coders/observers
Consistency among multiple indicators
Estimates of Reliability
Statistical coefficients that tell use how
consistently we measured something
Powerpoint Templates
Measurement Validity
Are we really measuring concept we defined?
Is it a valid way to measure the concept?
Many different approaches to validation
Judgmental as well as empirical aspects
Powerpoint Templates
Key to Reliability and Validity
Concept explication
Thorough meaning analysis
Conceptual definition:
Defining what a concept means
Operational definition:
Spelling out how we are going to measure
concept
Powerpoint Templates
Four Aspects of Reliability:
1. Stability
2. Reproducibility
3. Homogeneity
4. Accuracy
Powerpoint Templates
1. Stability
Powerpoint Templates
2. Reproducibility
Powerpoint Templates
3. Homogeneity
Powerpoint Templates
4. Accuracy
Powerpoint Templates
Increasing Reliability
General:
Training coders/interviewers/lab personnel
More careful concept explication (definitions)
Specification of procedures/rules
Reduce subjectivity (room for interpretation)
Survey measurement:
Increase the number of items in scale
Weeding out bad items from item pool
Content analysis coding:
Improve definition of content categories
Eliminate bad coders
Powerpoint Templates
Indicators of Reliability
Test-retest
Make measurements more than once and see if they
yield the same result
Split-half
If you have multiple measures of a concept, split
items into two scales, which should then be correlated
Cronbachs Alpha or Mean Item-total Correlation
Powerpoint Templates
Konsistensi Dalaman
Satu ujian diambil hanya sekali
Mengukur sejauh mana item dalam satu ujian
adalah konsisten antara satu sama lain dan
dengan keseluruhannya
Terdiri daripada tiga jenis:
1. Kebolehpercayaan belah dua (split-half reliability)
2. Kebolehpercayaan Kuder-Richardson20
3. Kebolehpercayaan Cronbachs alpha
Powerpoint Templates
Kebolehpercayaan Split-
Half
Melibatkan pembahagian satu ujiankepada dua sub ujian
Setiap ujian mengandungi sebahagian()daripada
bilangan item yang asal
Tujuan membahagikan kepada dua subujian ialah untuk
mendapatkan dua ujian yang selari mungkin
Kaedah ini sesuai untuk satu ujian yang panjang dan
mengambil masa yang lama atau susah untuk diuji
Prosedur
Jalankan keseluruhan ujian terhadap satu kumpulan
Bahagikan ujian tersebut kepada duabentuk yang berbeza
dengan memilih item ganjil atau item genap
Kirakan setiap skor pelajar bagi kedua-dua bahagian
Hitungkan pekali korelasi bagi kedua-dua skor
Aplikasikan formula pembetulan Spearman-Brown
Menilai keputusan Powerpoint Templates
SPLIT-HALF RELIABILITY
(Spearman-Brown)
Reliability = 2r
1 r
r = the actual correlation between the two halves
of the instrument (e.g. 0.85);
2 ( 0.85 ) 1.70
Reliability = = = 0.919
1 0.85 185
Powerpoint Templates
Kuder-Richardson 20 (KR
20)
Rumus tersebut dibuat berdasarkan penskoran
secara dikotomi (betul-salah) seperti dalam ujian
objektif pelbagai pilihan.
Formula KR 21 lebih mudah dihitung dan
mengurangkan masa pengiraan tetapi kurang tepat
berbanding KR 20.
Powerpoint Templates
Kuder-Richardson 20 (KR 20)
Powerpoint Templates
Powerpoint Templates
Kebolehpercayaan Cronbachs
alpha
Digunakan apabila ujian mempunyai skor
yang lebih daripada dua pilihan
Menganalisis konsisten dalaman jika
nombor digunakan untuk menunjukkan
pilihan respon
Pekali alpha yang kerap kali digunakan
dalam menganggarkan kebolehpercayaan
ketekalan-dalaman
Powerpoint Templates
Kebolehpercayaan Skorer
(Scorer Reliability)
Mengkaji individu yang mendapat skor untuk ujian yang
dijalankan
Interjudge- merujuk kepada dua atau lebih individu
yangmendapat skor yang konsisten
Intrajudge- merujuk kepada individu yang mendapat skor
yang konsisten selepas satu jangka masa
Penting apabila penskoran berpotensi sebagai subjektif
Penting untuk menentukan kebolehpercayaan seseorang
individu yang menskor ujian terbuka (tidak terhad)(open-
ended tests)
Kebolehpercayaan penskor penting untuk ditentukan
apabila perlakuan ke atas ujian mempunyai kesan yang
serius terhadap pengambil ujian
Powerpoint Templates
Reliability and Validity
Powerpoint Templates
Reliable but not Valid
Powerpoint Templates
Reliable and Valid
Powerpoint Templates
Types of Validity
1. Face validity
2. Content validity
3. Pragmatic (criterion) validity
A. Concurrent validity
B. Predictive validity
4. Construct validity
A. Testing of hypotheses
B. Convergent validity
C. Discriminant validity
Powerpoint Templates
Face Validity
Powerpoint Templates
Content Validity
Subjective judgment of experts about:
what is not there
Start with conceptual definition of each
dimension:
Is it represented by indicators at the
operational level?
Are some over or underrepresented?
If current indicators are insufficient:
develop and add more indicators
Example--Civic Participation questions:
Did you vote in the last election?
Do you belong to any civic groups?
Have you ever attended a city council
meeting?
What about protest participation or online
organizing?
Powerpoint Templates
Pragmatic Validity
Powerpoint Templates
Construct Validity
Powerpoint Templates
A. Testing Hypotheses
Powerpoint Templates
B. Convergent Validity
Powerpoint Templates
C. Discriminant (Divergent) Validity
Powerpoint Templates
Dimensions of Validity for
Research Design
Internal
Validity of research design
Validity of sampling, measurement,
procedures
External
Given the research design, how valid are
Inferences made from the conclusions
Implications for real world
Powerpoint Templates
Internal and External Validity in
Experimental Design
Internal validity:
Did the experimental treatment make a
difference?
Or is there an internal design flaw that invalidates the results?
External validity:
Are the results generalizable?
Generalizable to:
What populations?
What situations?
Powerpoint Templates
Core Qualitative Methods
Powerpoint Templates
Guiding Principles
Qualitative research designs consider ways to foster:
Reflexivity (an ongoing process of reflecting on the researchers subjective experience,
ways to broaden and enhance this source of knowing, & examining how it informs research)
Powerpoint Templates
The Iterative Process of Qualitative
Research: A Model
Analysis
Data Collection
Powerpoint Templates
Reflection
Qualitative Research Techniques
Instrumentation:
Powerpoint Templates
Qualitative Research Techniques/
Considerations
Sampling
Powerpoint Templates
Qualitative Research Techniques
Data Collection
Powerpoint Templates
Qualitative Research Techniques
Analysis
Key Informant Feedback
Codebooks (specifies definitions and relationships of concepts and terms)
Memos (emerging patterns, concepts; documentation of analytic pathways)
Case Analysis Meeting (a meeting of a research team for the purposes of reflecting on analytic process,
tools, and findings)
Matrices or Diagrams (to identify and examine time sequencing, the structure of relationships, conditions of
cross case events)
Powerpoint Templates
Qualitative Research Techniques
Analysis
Key Informant Feedback
Codebooks (specifies definitions and relationships of concepts and terms)
Memos (emerging patterns, concepts; documentation of analytic pathways)
Case Analysis Meeting (a meeting of a research team for the purposes of reflecting on analytic process,
tools, and findings)
Matrices or Diagrams (to identify and examine time sequencing, the structure of relationships, conditions of
cross case events)
Powerpoint Templates
Three Qualitative Methods:
Ethnography
Focus: study human behaviour in the
cultural context in which it is embedded
Ethnography is the work of describing a
culture the way of life of a cultural group
Associated with Cultural Anthropology
Example: David Counts and Dorothy Ayer
Counts: An Ethnography of RVing
Seniors
Powerpoint Templates
Three Qualitative Methods:
Grounded Theory
Focus: develop a theory to explain underlying
social processes of a cultural group
Useful in areas where little is known or when a
new perspective is needed
Used for exploratory, descriptive studies
Because the theory emerges from the data, it is
said to be grounded in the data
Foundation in Symbolic Interactionism
Example: Kerry Daly and the social
construction of fatherhood
Powerpoint Templates
Three Qualitative Methods:
Phenomenology
Focus: reveal the meaning of the lived experience from
the perspective of participants
Describe the essences of lived experience
Essences: elements related to the true meaning of
something that gives common understanding to the
phenomenon under study
Conveyed with descriptive language
Drawn from Philosophy; used across disciplines
Example: J.E. Solchany: A phenomenological study of
womens preadoptive experiences
Powerpoint Templates
Data Collection Methods
in Qualitative Studies
Three data collection strategies
introduced:
1. Participant observation
2. In-depth interviews
3. Focus group interviews
Powerpoint Templates
Steps in Participant Observation:
A. Gaining Entry into the Group
Take into consideration the type of group
formal organizations require formal entry;
involves letter writing, permission requests,
etc.
Informal groups different strategy needed
Access may be gained through a gatekeeper
(an individual with special status)
Want to involve key informants (those who are
most knowledgeable about the group)
Powerpoint Templates
Steps in Participant Observation:
B. Developing/Maintaining Rapport
Researcher must work hard to develop
and maintain good relationships in the
field
E.g., be sure not to become associated
with one faction in a group or organization
Researcher could be blamed for
problems that arise in the setting
Powerpoint Templates
Steps in Participant Observation:
C. Strategies for Taking Field Notes
Field notes integral to participant observation
Include descriptions and interpretations of individuals,
interactions, and events
Distinguish descriptions from interpretations
Record time and location of observations, as well as
key information (weather, events happening and their
significance)
Keep theoretical memos which are the tentative
interpretations emerging and being assessed through
further data collection
Powerpoint Templates
Field Notes (contd)
Powerpoint Templates
Data Collection Methods:
3. Focus Group Interviews
Interview format, but in a group setting
6-12 participants with common experience
Dates back to the 1940s used to assess
effectiveness of morale-boosting radio shows
1970s onward used by market researchers
1980s onward used by academics
Transcript of discussion is the data
Plus accompanying notes
Use content analysis or grounded theory
approach to analyze the data
Powerpoint Templates
3. Focus Group Interviews (contd)
Strengths:
Open-ended question
Spontaneously deal with issues as they arise
Cost-effective method of collecting data
Less time-consuming
Weaknesses:
One or two participants may dominate
Not done in a natural setting, so little observation to
help understand the experience of the participants
Powerpoint Templates
Implications for Handling Threats
to Validity and Reliability
In quantitative research, threats to
validity are addressed by prior design
features (such as randomization and
controls)
Powerpoint Templates
Implications for Handling Threats
to Validity and Reliability
In qualitative research, such prior elimination of threats to
validity is less possible because:
qualitative research is more inductive, and
Powerpoint Templates
Validity
Powerpoint Templates
Assessing the Validity of
Qualitative Research
Can another research read your field (and other types of) notes
(i.e., the explication of your logic) and come to the same
understandings of a given phenomenon?
Concern about validity (as well as reliability) is the primary
reason thick description is an essential component of the
qualitative research enterprise
Resources:
Handout: Different Types of Notes
Powerpoint Templates
Major Types of Validity in
Qualitative Research
Descriptive Validity
Interpretive Validity
Theoretical Validity
Powerpoint Templates
Descriptive Validity
Powerpoint Templates
Descriptive Validity
Powerpoint Templates
Interpretive Validity
Powerpoint Templates
Interpretive Validity: Design
Consideration
While the relevant consensus about the terms used in
description rests in the research community, the
relevant consensus for the terms used in
interpretation rests, to a substantial extent, in the
community studied
An important design element, for increasing
interpretive validity, therefore, is to employee, at
some level/to some degree, a participatory research
approach (e.g., through member checks, peer to peer
research model, etc.)
Powerpoint Templates
Theoretical Validity
Powerpoint Templates
Major Threats to Validity
Type I error: believing a principle to
be true when it is not (i.e., mistakenly
rejecting the null hypothesis)
Type II error: rejecting a principle
when, in fact, it is true
Type III error: asking the wrong
question
Powerpoint Templates
Triangulation: An Important
Theoretical Validity Check
Case example: Parable of the blind men
and the elephant
Powerpoint Templates
Triangulation: An Important
Theoretical Validity Check
The most fertile search for validity comes from a
combined series of different measures, each with its
own idiosyncratic weaknesses, each pointed to a
single hypothesis. When a hypothesis can survive
the confrontation of a series of complementary
methods of testing, it contains a degree of validity
unattainable by one tested within the more
constricted framework of a single method.
Powerpoint Templates
Multi-site Studies: Another Way to
Enhance Generalizability
A finding emerging repeatedly in the study of
numerous sites would appear to be more likely to be
a good working hypothesis about some as yet
unstudied site than a finding emerging from just one
or two sites.
A finding emerging from the study of several very
heterogeneous sites would be more robust and, thus,
more likely to be useful in understanding various
other sites than one emerging from the study of
several very similar sites.
Heterogeneity may be obtained by creating a
sampling frame that maximizes the variation inherent
in the sample, specifically in terms of potentially
Powerpoint Templates
theoretically important dimensions
Reliability
Powerpoint Templates
Enhancing the Reliability of
Qualitative Research
Researchers can enhance the reliability of their
qualitative research by:
Standardizing data collection techniques and protocols
Powerpoint Templates