Psychometrics: An Introduction

Citation

Furr, R. M. (2021). Psychometrics: An Introduction (4th ed.). SAGE Publications.

Chapter Summary

The fourth edition has fourteen chapters. An opening chapter is followed by five parts: basic concepts in measurement (Chapters 2–4), reliability (5–7), validity (8–9), threats to psychometric quality (10–11) and advanced psychometric approaches (12–14). The publisher describes the orientation as conceptual: the book concentrates on purpose rather than procedure, on the "why" rather than the "how," and favors concepts, logic and practical applications over mathematical proofs. Most chapters end with a technical appendix of R syntax, a summary and suggested readings.

Chapter 1: Psychometrics and the Importance of Psychological Measurement

The introduction explains why psychological testing matters, how observable behavior relates to unobservable psychological attributes, what psychological tests are and what types exist, and what psychometrics is. It also covers the challenges of measurement in psychology and the importance of individual differences, with the point that psychometrics goes well beyond differential psychology.

Part I, Chapter 2: Scaling

This chapter starts from basic questions about numbers: units of measurement, additivity and counting, and the four scales of measurement. It then considers the practical implications of these scales and some additional issues about them.

Part I, Chapter 3: Differences, Consistency, and the Meaning of Test Scores

The chapter treats variability and distributions of scores, ways of quantifying association or consistency between distributions, variance and covariance for composite variables, and binary items. It closes with interpreting test scores and test norms.

Part I, Chapter 4: Test Dimensionality and Factor Analysis

The chapter asks how many dimensions a test measures and introduces factor analysis as the tool for examining test dimensionality.

Part II, Chapters 5–7: Reliability: Conceptual Basis; Empirical Estimates of Reliability; The Importance of Reliability

Chapter 5 sets out classical test theory: observed scores, true scores and measurement error, the variances of each, four ways to think about reliability, the standard error of measurement, and domain sampling theory. Chapter 6 covers empirical estimation by alternate forms, test–retest and internal consistency methods, together with sample heterogeneity, reliability generalization and the reliability of difference scores. Chapter 7 shows why reliability matters in applied practice, in behavioral research and in test construction and refinement.

Part III, Chapters 8–9: Validity: Conceptual Basis; Estimating and Evaluating Convergent and Discriminant Validity Evidence

Chapter 8 defines validity and organizes validity evidence by source: test content, internal structure, response processes, associations with other variables and consequences of testing; it also treats other perspectives on validity and contrasts reliability with validity. Chapter 9 concentrates on the nomological network and on methods for evaluating convergent and discriminant evidence, the factors that affect a validity coefficient and how to interpret one.

Part IV, Chapters 10–11: Response Biases; Test Bias

Chapter 10 covers types of response bias and ways of coping with them, including response sets and styles. Chapter 11 examines why score bias matters, how to detect construct bias and predictive bias, other statistical procedures and test fairness, and uses the SAT as a worked example.

Part V, Chapters 12–14: Confirmatory Factor Analysis; Generalizability Theory; Item Response Theory and Rasch Models

Chapter 12 returns to factor analysis in its confirmatory form, applied to a scale's internal structure, reliability, validity and measurement invariance. Chapter 13 introduces generalizability theory, with G studies and D studies and one-facet and two-facet designs. Chapter 14 introduces item response theory and Rasch models, including a one-parameter example, model fit, item and test information, and applications.

Key Concepts

Part I: Basic concepts in measurement

• Scales of measurement: the four scales and what each implies for how scores can be treated.
• Test norms: reference distributions used to interpret an individual's score.
• Test dimensionality: whether a test measures one attribute or several, examined with factor analysis.

Part II: Reliability

• Observed, true and error scores: the classical decomposition on which reliability is defined.
• Standard error of measurement: how far an observed score may stray from the true score.
• Reliability estimates: alternate forms, test–retest and internal consistency approaches.

Part III: Validity

• Sources of validity evidence: content, internal structure, response processes, associations with other variables and consequences of testing.
• Nomological network: the web of expected relations used to judge convergent and discriminant evidence.

Parts IV–V: Threats and advanced methods

• Response biases and test bias: threats to the quality and fairness of scores.
• Confirmatory factor analysis, generalizability theory, item response theory: the three advanced frameworks the book introduces.

Critical Analysis

Strengths

• Conceptual orientation stated plainly. The publisher's description makes the book's aim explicit, purpose over procedure and understanding over proof, and the section headings bear this out: they are organized around questions such as "What is validity?" and "Why worry about test score bias?"
• Statistics introduced where needed. The publisher notes that statistical procedures are introduced in the context of their use, not in a separate chapter, so readers meet variance and covariance when they need them for scores and reliability.
• R support in the fourth edition. Technical appendices in R at the end of most chapters, plus dataset and syntax files, let readers practice the computations. The publisher also lists expanded figures and tables, and additional coverage of fundamental statistics.
• Clear sequence. Reliability and validity each get a conceptual chapter before the empirical chapters, and the advanced material comes last.

Limitations

• Advanced methods get one chapter each. Confirmatory factor analysis, generalizability theory and item response theory each occupy a single chapter, so they are introductions; readers needing practice in latent-variable modeling or IRT will need a specialist text.
• Conceptual by design. Readers seeking proofs or derivations will not find them, since the book deliberately favors concepts over mathematical proofs.
• Focus on psychological measurement. Examples are drawn from behavioral science research, so educators will need to translate the ideas to classroom and large-scale educational tests.

Where It Fits in the Course

In the assessment course, Furr is the measurement-theory book. Popham's Classroom Assessment and Brookhart and Nitko's Educational Assessment of Students are teacher-oriented texts that explain validity and reliability and then show how to build and use classroom assessments; they are the place to start for practice. Wiliam's Embedded Formative Assessment is a practical account of formative assessment, and the Handbook of Formative Assessment in the Disciplines is a research handbook that links formative assessment to measurement principles in particular subjects.

Furr's book is the one to add when the question shifts from how to assess to how good a score is: what reliability and validity mean quantitatively, how bias is detected, and how modern approaches such as item response theory extend classical test theory. It pairs naturally with the handbook, whose editors connect formative assessment with measurement principles.

How to Use This Book

The book suits graduate students in education and psychology, and practitioners who build, select or interpret tests. A course in classroom assessment and an introductory statistics course covering correlation and variance are the sensible preparation.

A practical plan:

• Core: Chapters 1–3 and 5–6 for the vocabulary of scores, variance and reliability, then Chapters 8–9 for validity.
• Applied: Chapter 7 on why reliability matters in practice and Chapter 11 on test bias for anyone choosing or reviewing tests.
• Optional: Chapter 4 and Chapter 12 together for factor analysis; Chapters 13 and 14 for generalizability theory and item response theory after the classical material is secure.

Work through the R appendices on the supplied datasets while reading, and apply each idea to a test you know, asking what reliability evidence and what validity evidence exist for it.