This is a slideshow for my first self-published book on survey research replication. In this slideshow, I summarize my book's eight chapters. The slideshow is navigable and works best in Office 365 (due to use of Zoom).
2. Quick Facts
First book authored by
James M. Tamayose, M.Ed.
Written in American Psychological
Association 6th ed. Writing Style
238 references, 13 tables,
and 16 images
3. A couple of notes before going into the synopsis
My book represents my abilities and limitations
at the time I wrote my book.
My imperfections are special to me because
I can expand and contract from them.
11. Chapter 1. Book Overview (can be read on Amazon)
1. Book
Structure
3. My
Book’s
Benefits
2. The
Need for
My Book
4. Why I
Chose
Replication
5. Why
Me
as Author
13. Researchers cannot effectively conduct replication
studies without knowing replication’s definitions.
2.1 Replication’s Definition
Before researchers decide what to replicate,
they have to know what they are about to do.
14. Replication’s multiple definitions inspire
multiple kinds of replication studies
2.2 Research Replication Kinds
Note:
Section 2.2 has five tables describing different kinds of replication research
15. Replication is deeper than
establishing confidence in research
2.3 Why do Researchers Care?
16. Aside from replicating things from earlier studies,
other barriers to replication exist.
2.4 Replication is NOT Easy
18. Surveys are a vehicle for social cognition, the
study of how people understand the self, others,
and the interdependence between self and others
(Fiske & Taylor, 2013; Markus & Kitayama, 1991).
3.1 Social Cognition
Fiske, S.T. & Taylor, S.E. (2013). Social Cognition: From brains to culture (2nd ed.).
Thousand Oaks, CA: Sage.
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for
cognition, emotion, and motivation. Psychological Review, 98(2), 224–253.
19. Researchers cannot effectively conduct replication
studies without knowing replication’s definitions.
3.2 Face Validity and Pretesting
20. William Shakespeare wrote
“The Taming of the Shrew”
o True
o False
William Shakespeare wrote
“The Taming of the Shrew”
o False
o True
Version #1 Version #2
Attaining face validity: Depends on what is being studied and
how well questions appear to measure whatever is being studied.
Version #1 and Version #2 could establish face validity for a test
on Shakespearean Literature.
21. William Shakespeare wrote
“The Taming of the Shrew”
o True
o False
Version #1
Version #2
Pre-Testing: Pre-study quality control, useful for getting
feedback.
Pre-testing Version #2 may elicit negative feedback from people
accustomed to seeing True listed as the first option.
For the formal survey administration, Version #1 may be used.
William Shakespeare wrote
“The Taming of the Shrew”
o False
o True
Version #2
22. Wording differences can yield differing responses
even when questions’ substantive content are the
same (Schumann & Presser, 1977, 1986).
3.3 Word Choice
Schuman, H., & Presser, S. (1977). Question wording as an independent variable in
survey analysis. Sociological Methods and Research, 6(2), 151–170.
Schuman, H., & Presser, S. (1996). Questions and answers in attitude surveys:
Experiments in question form, wording, and content. Thousand Oaks, CA: Sage.
23. 3.4 Item Bias
Biased Items are items with differing
psychological meanings functioning differently
across groups (Ackerman, 1992; Holland &
Thayer, 1988; van de Vijver & Tanzer, 1997).
Ackerman, T. A. (1992). A didactic explanation of item bias, item impact, and item validity from a
multidimensional perspective. Journal of Educational Measurement, 29(1), 67–91.
Holland, P. W., & Thayer, D. T. (1988). Differential item performance and the Mantel-Haenszel procedure.
In H. Wainer & H. I. Braun (Eds.), Test validity (pp. 129–145). Hillsdale NJ: Erlbaum.
van de Vijver, F. J. R., & Tanzer, N. K. (1997). Bias and equivalence in cross-cultural assessment: An
overview. European Review of Applied Psychology, 47(4), 263–279.
24. A response combination is a specific way a
respondent responds to a set of items.
Response combinations differ conceptual from
response styles and response sets.
3.5 Response Combinations
From Peer and Gamliel (2011), response style, tendency to distort responses in a certain direction;
response set, respondents’ desire to express a certain image to researcher(s).
Peer, E., & Gambiel, E. (2011). Too Reliable to Believe? Response Bias as a Potential Source of
Inflation in Paper-and-Pencil Questionnaires Reliability. Practical Assessment, Research and
Evaluation. 16(9).
26. If false null hypotheses testing for significance is
bad, then imagine the replicating of false null
hypotheses for significance.
4.1 Null Hypotheses
For surveys, false null hypotheses may be easy to set up
because survey items cannot be truly isolated from the
survey context.
27. Researchers cannot effectively conduct replication
studies without knowing replication’s definitions.
4.2 Thoughts on Bayesian Stats
28. Creative uses of ransacking, partitioning, and the
Yates’ Correction may improve data analyzability
4.3 Contingency Tables
Key References used for Section 4.3
Fisher, R. A. (1954). Statistical methods for research workers. Edinburgh, UK: Hafner Publishing.
Goodman, L. A. (1969). How to ransack social mobility tables and other kinds of cross –
classification tables. American Journal of Sociology, 75(1), 1–40.
Sharpe, D. (2015). Your Chi-Square Test is Statistically Significant: Now What? Practical
Assessment, Research & Evaluation, 20(8), 1–10.
Yates, F. (1934). Contingency tables involving small numbers and the χ 2 test. Supplement to the
Journal of the Royal Statistical Society, 1(2), 217–235.
29. 4.4 Broken Latent Variables
Fundamentally, researchers cannot analyze latent
variables (e.g. factors and classes) that researchers
cannot describe or explain.
In Section 4.4, I describe five ways (e.g. item wording and its
implications; response combinations exceeding population size)
that latent variables become broken.
30. A response combination is a specific way a
respondent responds to a set of items.
Response combinations differ conceptual from
response styles and response sets.
4.5 Missing Data
31. A Multiple Imputation Shortcoming?
Survey with
missing data and
unobserved
response
combinations
Response combinations
that were unobserved
prior to multiple
imputation to become
observed
Multiple imputation
may result in
Two kinds of observed response combinations:
(1) Observed without imputation and (2) Observed because of imputation
Are these two kinds of response combinations comparable?
32. A Maximum Likelihood Estimation Issue?
Maximum likelihood estimation works around missing data
(i.e. uses available data but does not lead to imputed values)
Do the two tables of response combinations below differ?
True-True True-False
False-True False-False
True-True True-False True-Miss
False-True False-False False-Miss
Miss-True Miss-False Miss-Miss
34. Contains three short survey-related institutional
research scenarios.
Chapter 5. Institutional Research
Short highlight: Grey areas of high stakes institutional
research survey findings’ meaning(s) can be easily maneuvered
and difficult to understand when replicated.
35. Consists of two scenarios looking at different
survey research-related practices in mental health
Chapter 6: Mental Health
Short highlight 1: Specific response combinations are powerful
Short highlight 2: Item parceling can be a dangerous practice
36. Composed of coverage on equivalence and an
examination into Dr. Morris Rosenberg’s Self-
Esteem Scale (RSES)
Chapter 7: Cross-cultural Psychology
Short highlight 1: Equivalence is not just statistical, it’s
conceptual as well
Short highlight 2: Cross-cultural equivalence is hard to
establish, replicating it is even harder.
37. The Gems are a collection of valuable thoughts that I
could not fit into one of the other chapters.
In my book’s earlier drafts, I incorporated The Gems
into the text but doing so resulted in mixing scholarly
and anecdotal content.
For myself, keeping The Gems separate taught me
about writing efficiency.
Chapter 8. The Gems