Test Theory for a New Generation of Tests

Test Theory for a New Generation of Tests

Test Theory for a New Generation of Tests

Test Theory for a New Generation of Tests


The editors of this volume suggest that there are missing elements in the conceptualization upon which standard test theory is based. Those elements are models for just how people know what they know and do what they can do, and the ways in which they increase these capacities. Different models are useful for different purposes; therefore, broader or alternative student models may be appropriate. The chapters in this volume consider a variety of directions in which standard test theory might be extended. Topics covered include: the role of test theory in light of recent work in cognitive and educational psychology, test design, student modeling, test analysis, and the integration of assessment and instruction.


Robert J. Mislevy Educational Testing Service

A full century has passed since papers on statistical test theory first began to appear (Edgeworth, 1888, 1892; Spearman, 1904a, 19Nb, 1907, 1910, 1913). In his 1961 review of Measurement of Learning and Mental Abilities, Gulliksen (1961) succinctly characterized the field as follows:

The central problem of test theory is the relation between the ability of the individual and his [or her] observed score on the test . . . Psychologists are essentially in the position of Plato's dwellers in the cave. They can know ability levels only through the shadows (the observed test scores) cast on the wall at the back of the cave. The problem is how to make the most effective use of these shadows (the observed test scores) in order to determine the nature of reality (ability) which we can know only through these shadows. (p. 101)

Following the aforementioned work, a sequence of influential and increasingly mathematically sophisticated books followed, notable among them Thorndike (1919) An Introduction to the Theory of Mental and Social Measurements, Kelley's (1927) Interpretation of Educational Measurements, Guilford (1936) Psychometric Methods, Gulliksen (1950/1987) own Theories of Mental Tests, Rasch (1960/1980) Probabilistic Models for Some Intelligence and Attainment Tests, and Lord and Novick (1968) encyclopaedic Statistical Theories of Mental Test Scores, which includes Birnbaum's contributions on item response theory. Why, now, Test Theory for a New Generation of Tests? A point of departure from Gulliksen's description portends our motivation.

We would concur with Gulliksen that the heart of test theory is connecting . . .

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.