Medicine > Psychiatry, Neurology > Statistics in Psychosocial Research: Measurement

Statistics in Psychosocial Research: Measurement posted by boym on 3/24/2008

Add To Favorites

Abstract/Syllabus

Courseware/Lectures

Test/Tutorials

Staff

Instructors: William Eaton, Elizabeth Garrett-Mayer, and Jeannie-Marie Leoutsakos

Originally Offered

Fall 2006

Offered By

Department of Mental Health

Description

Presents quantitative approaches to measurement in the psychological and social sciences. Topics include the principles of psychometrics, including reliability and validity; the statistical basis for latent variable analysis, including exploratory and confirmatory factor analysis and latent class analysis; and item response theory. Draws examples from the social sciences, including stress and distress, social class and socioeconomic status, personality; consumer satisfaction, functional impairment and disability, quality of life, and the measurement of overall health status. Intended for doctoral students.

OCW offers a snapshot of the content used in courses offered by JHSPH. OCW materials are not for credit towards any degrees or certificates offered by the Johns Hopkins Bloomberg School of Public Health.

Schedule

SESSION # TOPIC ACTIVITIES

1
Introduction to Measurement

After this class students will be able to (1) briefly describe the concept of reliability in both intuitive and statistical terms, and (2) identify the key assumptions of classical test theory.

Introduction

Syllabus review

Classical test theory

Introduction to reliability

2
Measuring Association and Dimensionality

After this class students will be able to (1) measure associations between continuous observed variables using covariances and correlations, (2) measure magnitudes of association between discrete observed variables, and (3) define multidimensionality regarding latent variables.

Covariance

Pearson & Spearman correlation

Correlations with non-linear data

Polychoric correlation

Covariance, correlation, and odds ratio matrices

Dimensionality

3
Principles of Psychometrics: Reliability I

After this class students will be able to (1) describe two definitions of the concept of reliability, (2) predict how long a scale should be, and (3) estimate reliability for continuous and categorical measures

Types of reliability (Interrater, test-retest reliability, internal consistency)

Different types of reliability coefficients (correlation, split half measures, Alpha coefficient, Kuder Richardson Coefficient, Kappa

4
Principles of Psychometrics: Reliability II

After this class students will be able to (1) describe the relationship of the intraclass correlation coefficient to other measures of reliability, and (2) correctly identify which intraclass correlation to use for different research designs

ANOVA model for reliability

Intra-class Correlation Coefficient

Research Designs

5
Principles of Psychometrics: Validity I

After this class students will be able to (1) distinguish four different types of validity, (2) describe the conceptual and quantitative relationship of reliability to validity, (3) estimate a true correlation from an observed correlation

Types of Validity (face, content, criterion, construct)

Relationship of Reliability to Validity

Correction for attenuation

6
Principles of Psychometrics: Validity II

After this class students will be able to evaluate the relative utility of different cutoffs for a measure in relation to a gold standard.

Internal Construct Validity

External Construct Validity

Multi-trait Multimethod Matrix

Sensitivity and Specificity

ROC Curves

7
Scale Development

After this class students will be able to describe procedures for constructing a scale from scratch
Scale Construction

8
Factor Analysis I

After this class students will be able to (1) identify when a factor analysis is appropriate and when it is not, (2) run a one-factor and multi-factor analysis, (3) interpret the results from a factor analysis

Introduction to factor analysis

The orthogonal factor model

Loadings

Principal components

Eigenvalues

Introduction to rotation

Communalities/Uniqueness of items

9
Factor Analysis II

After this class students will be able to (1) use the statistical procedure of rotation to aid in the interpretation of results from a factor
analysis, (2) be able to apply both orthogonal and oblique rotations and identify the assumptions
underlying each, and (3) apply the appropriate method of estimation for factor analysis

Factor extraction

Methods of estimation and rotation: orthogonal and oblique

Choosing the number of factors

Factor scores

Confirmatory factor analysis

Conditional independence Dichotomous factor analysis

10
Factor Analysis III

In this class students will (1) apply factor analysis to real data, and (2) critique published use of factor analysis
Journal examples

11
Latent Class Analysis I

After this class students will be able to (1) differentiate when to use factor analysis and when to use latent class analysis, and (2) interpret output from a latent class analysis

The latent class model

The response pattern matrix

Choosing the number of classes

Conditional probabilities

Interpreting the model

Examples: depression and functioning

12
Latent Class Analysis II

After this class students will be able to (1) estimate a latent class analysis, and (2) to interpret different criterion to choose among alternative models

Statistical model and assumptions

Exploration: response patterns

Issues of model fitting

Identifiability

Checking the model: tests and displays

13
Latent Class

In this class students will (1) apply latent class analysis to real data, and (2) critique published use of latent class analysis
Journal Examples

14
Sample Size in Reliability and Factor Analysis

After this class students will be able to (1) estimate the sample size needed to for scales with targeted reliability levels, (2) estimate the sample size needed for pilot studies that will use factor analysis, and (3) understand issues involved with sample size in latent class analysis.

Targeted Reliability Levels

Pilot Studies

Latent Class Analysis

15 Review

16 Final Exam

Readings

Readings are shown below. Click here for a printer-friendly version.

Textbooks

Required

Richard G. Netemeyer, William O. Bearden, Subhash Sharma. Scaling Procedures: Issues and Applications. Thousand Oaks, CA: Sage Publications, 2003.

A.L. McCutcheon. Latent Class Analysis. Newbury Park: Sage Publications, 1987.

Highly recommended

Robert F. DeVellis. Scale Development: Theory and Applications. Newbury Park: Sage Publications, 1991.

Jae-on Kim and Charles W. Mueller. Factor Analysis: Statistical Methods and Practical Issues. Beverly Hills, CA: Sage Publications, 1978.

SESSION #	TOPIC	READINGS

1	Intorduction to Measurement	No Reading
2	Measuring Association and Dimensionality	Required: Netemeyer. Pages xiii-xiv and 1-40. Recommended: Pagano, Gauvreau. Correlation. In: Principles of Biostatistics. Belmont, CA: Duxbury Press, 1993; 363-378. Digby PGN. Approximating the tetrachoric correlation coefficient. Biometrics 1983;39:753-757.
3	Principles of Psychometrics: Reliability I	Required: Netemeyer. Pages 41-59. Recommended: DeVellis. Pages 18-42 Anastasi A. Reliability. In: Psychological Testing , 6th edition. New York : Macmillan, 1988. Carmines EG, Zeller RA. Reliability and Validity Assessment . Beverly Hills, CA: Sage, 1979. Bohrnstedt G. Measurement. In: Rossi PH, Wright JD, Anderson AB, eds. Handbook of Survey Research. Orlando , FL : Academic Press, 1983. Shrout PE , Fleiss JL. Intraclass Correlations: uses in assessing rater reliability. Psychological Bulletin 1979;86:420-428.
4	Principles of Psychometrics: Reliability II	No Reading
5	Principles of Psychometrics: Validity I	Required: Netemeyer. Pages 71-94. Recommended: DeVellis. Pages 43-50. Anastasi A. Validity: basic concepts. In: Psychological Testing , 6th edition. New York, Macmillan, 1988. Carmines EG, Zeller RA. Reliability and Validity Assessment. Beverly Hills, CA: Sage, 1979. McCrae RR, Costa PT. Validation of the five-factor model of personality across instruments and observers. Journal of Personality and Social Psychology 1987;52:81-90. Bohrnstedt G. Measurement. In: Rossi PH, Wright JD, Anderson AB, Eds. Handbook of Survey Research, Orlando, FL: Academic Press, 1983.
6	Principles of Psychometrics: Validity II	No Reading
7	Scale Development	Required: Netemeyer. Pages 94-107. Streiner DL, Norman GR. From items to scales. In: Health Measurement Scales: A Practical Guide to Their Development and Use. 2nd edition. New York: Oxford University Press, 1995; 96-102. Murphy JM, et al. Performance of screening and diagnostic tests. Archives of General Psychiatry 1987;44:550-555.
8	Factor Analysis I	Required: Netemeyer. Pages 115-170. Recommended: DeVellis. pages 91-109. Kim J-O, Mueller CW. Factor Analysis: Statistical Methods and Practical Issues, Beverly Hills, CA: Sage Publications, 1988. Everitt B, Dunn G. Factor analysis. In: Applied Multivariate Data Analysis. London: Arnold Press, 2001.
9	Factor Analysis II	Recommended: Everitt B, Dunn G. Principal components analysis. In: Applied Multivariate Data Analysis. London: Arnold, 2001. Fisher LD, Van Belle G. Principal component analysis and factor analysis. In: Biostatistics: A Methodology for the Health Sciences, New York: John Wiley & Sons, 1993; 692-762. Johnson RA, Wichern DW. Principal components analysis (Chapter 8) and Factor Analysis and Inference for Structured Covariance Matrices (Chapter 9) In: Applied Multivariate Statistical Analysis, 2nd edition. Englewood Cliffs, NJ: Prentice Hall, 1988. Long JS. Confirmatory Factor Analysis: A Preface to LISREL. Beverly Hills, CA: Sage, 1983. Muthen B. A general structural equation model with dichotomous, ordered categorical, and continuous latent variable indicators. Psychometrika 1984;49:115-132. Eaton WW, Bohrnstedt GW, eds. Latent Variable Models for Dichotomous Outcomes: Analysis of Data from the NIMH Epidemiologic Catchment Area Program. Newbury Park: Sage Publications, 1989
10	Factor Analysis III	Required: Shapiro, Lasarev, McCauley. Factor Analysis of Gulf War Illness: What does it add to our understanding of possible health effects of deployment. Am J Epidemiol 2002;156:578-585. Lakka, Laaksonen, Lakka, Niskanen, Kumpausalo, Tuomilehto, Salonen. The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men. JAMA 2002;288:2709-2716. Hoodin, Kalbfleisch. Factor analysis and validity of the Transplant Evaluation Rating Scale in a large bone marrow transplant sample. Journal of Psychosomatic Research 2003;54:465-473. Costa PT, McCrae RR Four ways five factors are basic. In: Personality and Individual Differences 1992;13:653-665
11	Latent Class Analysis I	Required: McCutcheon A. Chapters 1 and 2. Latent Class Analysis , Newbury Park : Sage, 1987. Recommended: Uebersax J. LCA Frequently Asked Questions.
12	Latent Class Analysis II	Required: McCutcheon A. Chapter 3. In: Latent Class Analysis, Newbury Park, Sage, 1987.
13	Latent Class	Required: Nestadt, Addington, Samuels, Liang, Bienvenu, Riddle, Grados, Hoehn-Saric, Cullen. The Identification of OCD-Related Subgroups Based on Comorbidity. Biological Psychiatry 2003;53:914-920. Eaton WW, Dryman A, Sorenson A, McCutcheon A. DSM-III major depressive disorder in the community. A latent class analysis of data from the NIMH epidemiologic catchment area programme. British Journal of Psychiatry. 1989;155:48-54. Sullivan PF, Kessler RC, Kendler KS . Latent class analysis of lifetime depressive symptoms in the national comorbidity survey. Am J Psychiatry 1998; 155;1397-1406. Reboussin BA, Song E-Y, Shrestha A, Lohman KK, Wolfson M. A latent class analysis of underage problem drinking : Evidence from a community sample of 16-20 year olds. Drug and Alcohol Dependence 2006 ;83:199-209.
14	Sample Size in Reliability and Factor Analysis	Required: MacCallum RC, Widaman KF, Shaobo Z, Hong S. Sample size in factor analysis. Psychological Methods 1999;4:84-99.