Theory of measurement and evaluation; analysis of tests and scales emphasizing statistical and psychological analysis of experimental and standardized tests and scales. Pre: 416 or consent. (Cross-listed as PSY 616 and SW 658)
Modified: September 18, 2009
Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed). New York: McGraw-Hill.
Allen, M. J., & Yen, W. M. (1979). Introduction to measurement theory. Monterey, CA: Brooks/Cole.
Baker, F. (2001). The basics of item response theory. College Park, MD: ERIC Clearinghouse on Assessment and Evaluation, University of Maryland.
Crocker, L., & Algina, J. (1986). Introduction to classical and modern test theory. Wadsworth.
McDonald, R. P. (1999). Test theory: A unified treatment. Mahwah, NJ: Lawrence Erlbaum Associates.
Lord, F. M., & Novick, M. R. (1968). Statistical theories of mental test scores. Menlo Park, CA: Addison-Wesley.
Pedhazur, E. J., & Schmelkin, L. P. (1991). Measurement, design, and analysis: An integrated approach. Hillsdale, NJ: Elrbaum.
Students are strongly encouraged to make use of these Internet resouces/links as a way to reinforce understanding of basic concepts.
Research in the social, behavioral, educational, etc. sciences can be regarded as addressing questions with varying emphasis on the aspects of the inquiry process: specifying the domain of inquiry (i.e., ascertaining its relevant dimensions) and estimating the structural parameters of the theoretical domain that describe or "map" the domain.
This course, while focusing on measurement, is about modeling. An understanding of quantitative modeling depends on a conceptual integration of any list of the aspects of inquiry. No list, attempting to define the essence of research, can have on that list something that is independent of the other items thereon. Thus, no topic (such as psychometric/measurement theory) can be considered without a consideration of its relationship to the general logic of inquiry.
We will emphasize a consideration of the theories and techniques for dimensionalizing the domain but don't want to lose sight of the necessary integration with estimating the parameters of the structural model.
We will consider (primarily) variance/covariance structure of a matrix when the rows represent persons and the columns represent items. We will also consider the measurement model's relationship to theory (structural model).
I stole the following from an email msg posted to STAT-L by Eric Bohlman (ebohlman@netcom.com):
"H.G. Wells once said that he expected that in his lifetime, learning to think statistically would become as important as learning to read. Harold Hotelling once questioned how anyone could consider himself (sic) in possession of a liberal education if he (sic) didn't understand the nature of variation."
If you have difficulty with any of the concepts presented in the text or lectures, please consult the syllabus for one or more of the courses listed on my home page
Evaluation and assignment of grades
Questions or comments to: daniel@hawaii.edu