Robert DeVellis demystifies measurement by relating it to familiar experiences and by emphasizing a conceptual rather than a strictly mathematical understanding. Students' attention is drawn to important concepts that are foundational for subsequent topics, with opportunities provided to test understanding through chapter summaries and exercises. The Fourth Edition includes more attention to content validity and its relationship to scale breadth; a more thorough examination of coefficient alpha's limitations and remedies; discussion of "big measurement" vs "small measurement"; and additional discussion of the bifactor model in the chapter on factor analysis.
Prior to retiring in 2012, Robert F. DeVellis was Professor in the Department of Health Behavior, (Gillings School of Global Public Health) at the University of North Carolina at Chapel Hill. Dr. DeVellis has more than 35 years of experience in the measurement of psychological and social variables. He has been an active member of the Patient-Reported Outcomes Measurement Information System (PROMIS) consortium, a multisite National Institutes of Health (NIH) Roadmap initiative directed at identifying, modifying, testing, and disseminating outcome measures for use by NIH investigators. His role in PROMIS was as network-wide domain chair for Social Outcomes. He has served on the Board of Directors for the American Psychological Association's Division of Health Psychology (38), on the Arthritis Foundation's Clinical/Outcomes/Therapeutics Research Study Section, and on the Advisory Board of the Veterans Affairs Measurement Excellence Initiative. He is the recipient of the 2005 Distinguished Scholar Award from the Association of Rheumatology Health Professionals and is an associate editor of Arthritis Care and Research. In addition, he has served as guest editor, guest associate editor, or reviewer for more than two dozen other journals. He has served as principal investigator or co-investigator since the early 1980s on a series of research projects funded by the federal government and private foundations. He remains intellectually active in his editorial role and as a consultant on a variety of projects.
1. Overview General Perspectives on Measurement Historical Origins of Measurement in Social Science Later Developments in Measurement The Role of Measurement in the Social Sciences Summary and Preview Exercises 2. Understanding the Latent Variable Constructs versus Measures Latent Variable as the Presumed Cause of Item Values Path Diagrams Further Elaboration of the Measurement Model Parallel Tests Alternative Models Exercises Note 3. Reliability Methods Based on the Analysis of Variance Continuous versus Dichotomous Items Internal Consistency Reliability Based on Correlations Between Scale Scores Reliability and Statistical Power Generalizability Theory Summary Exercises Notes 4. Validity Content Validity Criterion-Related Validity Construct Validity Exercises 5. Guidelines in Scale Development Step 1: Determine Clearly What it is You Want to Measure Step 2: Generate an Item Pool Step 3: Determine the Format for Measurement Step 4: Have Initial Item Pool Reviewed by Experts Step 5: Consider Inclusion of Validation Items Step 6: Administer Items to a Development Sample Step 7: Evaluate the Items Step 8: Optimize Scale Length Exercises Note 6. Factor Analysis Overview of Factor Analysis Conceptual Description of Factor Analysis Bifactor and Hierarchical Factor Models Interpreting Factors Principle Components versus Common Factors Confirmatory Factor Analysis Using Factor Analysis in Scale Development Sample Size Conclusion Exercises 7. An Overview of Item Response Theory Item Difficulty Item Discrimination Guessing, or False Positives Item-Characteristic Curves IRT Applied to Multiresponse Items Conclusions Exercises 8. Measurement in the Broader Research Context Before Scale Development After Scale Administration Final Thoughts Exercises