Content area
Full text
The treatment of validity in the newest edition of Standards for Educational and Psychological Testing (Standards; American Educational Research Association, American Psychological Association, & National Council on Measurement in Education, 1999) is quite different from coverage in earlier editions of the Standards and in most measurement textbooks. The view of validity in the 1999 Standards is discussed, and suggestions/or instructors of measurement courses are offered.
In the newest edition of the Standards for Educational and Psychological Testing (Standards; American Educational Research Association [AERA], American Psychological Association [APA], & National Council on Measurement in Education [NCME], 1999), the treatment of validity is markedly different from what it was in the three earlier editions of the Standards (AERA, APA, & NCME, 1985;APA, 1954; APA, AERA, & NCME, 1966). The purpose of this article is to describe the meaning of validity in the new Standards, to compare that meaning with the presentation of validity in commonly used measurement textbooks, and to discuss implications for the teaching of validity in measurement courses for counseling students.
BACKGROUND
As Geisinger noted in 1992, the meaning of validity had undergone a "metamorphosis" during the previous half-century. In addition to Geisinger, other measurement experts and theorists who described historical changes in the definitions of validity include Angoff (1988), Cronbach (1988, 1989), Goodwin (1997, 2002a), Kane (1994, 2001), Messick (1988, 1989a, 1989b), Langenfeld and Crocker (1994), Moss (1992), and Shepard (1993). With the publication of the new Standards in 1999 (APA, AERA, & NCME, 1999), Geisinger's statement about the metamorphosis of validity definitions is as pertinent today as it was more than 10 years ago.
An early definition of validity, from the 1940s, emphasized the test itself. Validity was conceptualized as a static property of a measure-a view epitomized by Guilford's (1946) often cited statement that, "in a very general sense, a test is valid for anything with which it correlates" (p. 429). Other well-known psychometricians of the time who held the same general view of validity included Cureton (1951) and Gulliksen (1950). A test was considered to be either valid or not as evidenced by the correlations between the test and some other "external" criterion measure.
With the publication of the 1966 Standards (APA, AERA, & NCME, 1966), the meaning of validity...