| Part | Title | Key Chapters | | :--- | :--- | :--- | | | Introduction | Introduction | | Two | Statistical Foundations | Traditional Approaches to Scaling; Validity ; Elements of Statistical Description; Linear Combinations, Regression | | Three | Construction of Multi-Item Measures | The Theory of Measurement Error; The Assessment of Reliability ; Construction of Conventional Tests; Special Problems in Classical Test Theory | | Four | Factor Analysis | Recent Developments in Test Theory; Factor Analysis I: The General Model; Exploratory Factor Analysis II: Rotation; Confirmatory Factor Analysis | | Five | Additional Statistical Models | Profile, Discriminant Analysis, Multidimensional Scaling; Analysis of Categorical Data |
However, legal constraints apply, which we will address at the end of this article.
Originally published in 1967 and later updated with co-author Ira H. Bernstein, this text shaped how modern behavioral scientists approach measurement error, test construction, and statistical validation. The Significance of Nunnally’s Psychometric Theory
, which assesses how well individual items in a test correlate with one another to measure a single construct. psychometric theory nunnally pdf
While Nunnally’s text is the gold standard for Classical Test Theory, modern psychometrics has expanded significantly into Item Response Theory (IRT) and Generalizability Theory (G-Theory) . Researchers studying advanced testing environments (like computer-adaptive testing) often pair Nunnally's principles with newer literature focusing on IRT.
Digital textbook providers offer affordable e-book rentals of the text for research semesters.
Ideal for clinical, high-stakes decisions. | Part | Title | Key Chapters |
Reliability means the test is stable. [1, 4] A good test gives the same score every time. [1] If you weigh yourself twice, you should see the same number. Nunnally shows how to check for this stability. [1, 4]
Reliability alone is insufficient; a scale can be perfectly consistent but measure the entirely wrong thing. Nunnally outlined a comprehensive framework for —the degree to which evidence supports the interpretation of test scores.
For those who cannot obtain the PDF through legitimate means, several alternatives exist. The book is available for purchase in hardcover format through major retailers like eBay and ThriftBooks. Additionally, numerous graduate-level courses in psychometric theory—such as Psychology 405—list Nunnally and Bernstein (1994) as recommended reading, often providing selected chapters or supplemental materials through course websites. The Significance of Nunnally’s Psychometric Theory , which
: Are you actually measuring what you claim to be measuring?
Nunnally built his framework around Classical Test Theory, also known as the True Score Model. The basic mathematical equation is: X=T+Ecap X equals cap T plus cap E is the observed score (the score a person gets on a test).
Demonstrating that the test correlates with measures of theoretically similar constructs (convergent validity) and does not correlate with unrelated constructs (discriminant validity). Factor Analysis: The Heart of Test Construction
Academics, graduate students, and psychometricians frequently search for digital copies of this text for several key reasons: