PS505: Testing, Measurement and Assessment Dr. Raymond Brogan
December 16, 2014
Traditional Methods of Validity vs. the Process-Focused Model
Testing and assessments are used continuously in many areas of clinical psychology. The testing instrument needs to be considered reliable and valid. Cohen, Swerdlik and Struman (2013) define validity as being the estimate of how well the test measures what it is supposed to measure (p. 181). There are many different forms of validity that are used to assess the relevance of a specific test. The traditional forms of measurement used include criterion validity, construct validity, discriminant …show more content…
When the test designer creates these forms of testing, they have specific theories on ways the test taker will behave under certain conditions. If the test taker behaves according to the theory, then the test can be thought to show construct validity. The PF model, according to Bornstein (2011), states that validity is conceptualized by the rate at which the test taker reacts, as predicted, to a given variable (p. 536). The test administrator has the ability to manipulate certain variables to determine whether these manipulations are the reasoning for the predicted behaviors and can identify and manipulate extraneous variables as well to predict the behaviors that will or will not occur, which in the case of traditional methods, would be deemed challenging or inappropriate measures. Bornstein (2011) also states that construct validity can be separated into convergent validity and discriminant validity. Convergent validity is focused on the degree to what the test score is related to the variable being tested and discriminant validity is focused on the degree to which the test score is not related to the variable being …show more content…
The PF model, according to Bornstein (2011), hypothesizes the degree to which test takers can be shown to engage in a set of predictable “psychological processes” during the testing phase and, once these processes are identified, the degree to which experimental manipulation can be introduced and their impact on the test results (p. 536). Unlike traditional models of validity, which stray away from extraneous variables, the PF model embraces these confounds and looks at how they affect the test scores. Bornstein (2011) also states that the PF model compliments the traditional validity assessments in a way that will improve upon assessment procedures as well as enrich the researcher’s knowledge of testing bias and test scores (p.