William Revelle

Professor of Psychology
Department of Psychology
Northwestern University
Evanston, Illinois, USA 60208

Telephone: 847-491-7700
Office: 315 Swift Hall
Email: revelle@northwestern.edu

Contents

Psychology 405: Psychometric Theory

Objectives

  • 1) To acquaint you with the fundamental vocabulary and logic of psychological measurement and behavioral assessment.
  • 2) To develop your capacity for critical judgment of the adequacy of measures purported to assess behavior in the role of theory development.
  • 3) To acquaint you with some of the relevant literature in personality assessment, psychometric theory and practice, and methods of observing and measuring behavior.
  • 4) To instill in you an appreciation of and an interest in the principles and methods of psychometric theory in general and behavior assessment in particular.
  • 5) This course is not designed to make you into an accomplished psychometist (one who gives tests) nor is it designed to make you a skilled psychometrician (one who constructs tests), nor will it give you "hands on" experience with psychometric computer programs. Rather it is aimed to allow you to understand the fundamental theoretical issues concerning both the psychometrist and the psychometrician.
  • 6) Because modern psychometrics and statistics may be done using open source software such as R, examples will be presented in R. Instructions for installing and using R for psychometrics are available here.

Texts:

Sizable portions of several texts will be assigned.

  • Revelle, W. An introduction to psychometric theory with applications in R (under development). The chapters are available as pdfs. This is a draft and comments are appreciated.
  • Nunnally, Jum & Bernstein, Ira (1994) Psychometric Theory New York: McGraw Hill, 3rd ed.(recommended)
  • Loehlin, John (2004) Latent Variable Models: an introduction to factor, path, and structural analysis (4th Edition). Hillsdale, N.J.: LEA. ( recommended)

Requirements:

  • An objective exam sometime near the middle of the term.
  • An objective final exam
  • Practical experience in analyzing assessment instruments (done as homework and lab assignments).
  • A term project culminating in a paper < 12 pages long that reflects your specific area of expertise or interest within the larger framework of behavior assessment. Appointments will be scheduled during the term to discuss your general topic and progress towards its completion.

Course Outline

(last revision April 21, 2017)

  • Week
  • 1a


  • 1b


  • 2
  • 3a

  • 3b


  • 4

  • 5


  • 6
  • 7

  • 8
  • 9
  • 10
  • Topic
  • Correlational and Experimental psychology

  • Models of measurement


  • Variance and Covariance
  • Correlation and regression (continued)
  • Multiple Correlation and regression


  • Latent variable models

  • Latent variable models (2)


  • Classical Test Theory
  • Item Response Theory
  • Structural Equation Modeling
  • Validity + Reliability = SEM
  • Test Construction

See the old syllabus for more detail

405 Spring 2017 Outline:
Last revised April 11, 2017
Week Domain Readings Additional Course Notes Problem sets
Week 1I. IDs and experimental psychology Chapter 1: the role of measurement
Chapter 2 : Theory of data 		Overview
Theory of Data 		Getting started with R
homework #1 with answers
II. Models of measurement Chapter 3: The problem of scale metric properties and the problems of scale
also see the overall week 1 notes (above)		descriptive
homework with answers
Week 2
Correlation and regression (revised)		Variance/Covariance
Chapter 4: Correlation
Appendix D: Review of Matrix Algebra
Notes reviewing matrix algebra (revised)
John Fox's review of matrices 		Using R for statistics
An even shorter guide to R
Appendix-A: Using R
Appendix-B: R - commands
Chapter 3: The problem of scale		problem set 2 (using R)
the problems include the answers
More regression examples
Week 3 Correlation and regression (revised) Chapter 5 Multiple Correlation Regression using R
Multiple and partial correlation--alternative models
simple regression problem
answers
Week 4III. Latent variable models
Chapter 6: Constructs, Components and Factors

Factor analysis (revised notes)
Factor analysis (original notes) 		factor and principal components analysis
Examples of using R to do factor analysis
Simplex structures and item difficulty criteria for number of factors
clustering procedures
Week 6IV. Test theory
Chapter 7: Classical Test theory
Reliability (New)
Reliability (revised) (notes)
  • classical approaches
  • domain sampling
  • latent trait/item response
    		• More on reliability (revised)reliability hw
    reliability hw answers
    even more readable aswers
    Week 6B. Validity (predictive and construct) Midterm
    midterm sample
    Week 6validity + reliability = structural modelsC. Structural Models (conceptual introduction to causal models) Scoring a test (an example)
    Week 7D. Test Construction
    Week 8Personality Taxonomies
    The data box    		Structural Equation Modeling
    EFA and CFA
    causality?    		 Supplementary (excessive?) reading
    sem.chapter 3
    sem.chapter 4
    sem.chapter 5
    sem.chapter 6
    slides
    slides
    Week 9Summary of Psychometric theory
    Week 10Review

    405: Psychometric Theory: Class handouts

    Overview Also available as a pdf))

    Overview of the supplementary readings

    Prerequisites:

    Some basic familiarity with each of the following areas is highly recommended. Relevant readings are:

    How to view the reading list:

    This reading list can be seen as a general set of readings about psychometric theory and personality assessment. Not all of the readings are equally important. References in bold face are considered important and are well worth reading. The other readings are useful background if you want to explore a particular area in more depth. Except for Nunnally and Bernstein, the items on the reading list are not expected to be read for this course. Most of the items are meant to supplement your understanding of each particular area.

    In general, the more you understand Nunnally and Bernstein, the more you will understand the lectures. Similarly, the more you ask questions in class, the more you will understand Nunnally and Bernstein. The purpose of understanding Nunnally and Bernstein, of course, is that the more you understand the more you will be able to apply fundamental principles of measurement to your own research. It is the basic assumption of this course that all research can be improved by improving the quality of measurement used in that research.

    General Comment:

    Questions are encouraged. Remember, you probably are just as confused as your colleagues. Do not believe that you are the only person who does not understand. The overriding purpose of this course is to help you understand the fundamentals of measurement and personality assessment. Please ask questions if you do not understand. In the unlikely case that you are the only person who does not understand something, then see me after class or during office hours for further help.

    Some of the lectures are supplemented on the web with links from the syllabus. Also use the Psych Lit and the Social Science Citation Index search engines.

    General References

    Course Outline and Associated Readings

    I. Individual differences and experimental psychology: The role of assessment. II. Models of Measurement III. Test Theory and Practice