There is something about the topic of validity that seems to provoke dissatisfaction in many of those who encounter it--a sense that something is not right, and that something needs to be done to sort it out. Paul E. Newton in his target essay does not attempt a radical reconstruction of the validity edifice. His position is that the "consensus…

Paul E. Newton's article on the consensus definition of validity tackles a number of big issues and makes a number of strong claims. I agreed with much of what he said, and I disagreed with a number of his claims, but I found his article to be consistently interesting and thought provoking (whether I agreed or not). I will focus on three general…

This article responds to a critique of Ordinate Corporation's "Versant for English" test (formerly PhonePass and SET-10). The critique (Chun, 2006) purports to apply the language task framework of Bachman and Palmer (1996) and the Test Fairness framework of Kunnan (2004) as guides for an analysis. However, the analysis fails to apply either…

When causal inferences are to be synthesized across multiple studies, efforts to establish the magnitude of a causal effect should be balanced by an effort to evaluate the generalizability of the effect. The evaluation of generalizability depends on two factors that are given little attention in current syntheses: construct validity and external…

Comments on the article by B. Gawronski and G. V. Bodenhausen (see record 2006-10465-003). A metacognitive model (MCM) is presented to describe how automatic (implicit) and deliberative (explicit) measures of attitudes respond to change attempts. The model assumes that contemporary implicit measures tap quick evaluative associations, whereas…

Explores alternative empirical foundations of verbal and performance dichotomous constructs of Wechsler Intelligence Scales. Argues that Macmann and Barnett's position on the value of these constructs are based upon narrow interpretations of investigative results. Supports contention with IQ data from investigations of diverse populations. (CRR)