Adverse impact: Is it a viable concept?

It appears that those federal agencies responsible for enforcement persist in the belief that a given test score can mean one thing in one subgroup and something else in another and/or that inferences from a set of test scores can be valid for one subgroup but not for another. First surfacing in early agency regulations in connection with the now discontinued "differential validity" requirement, the belief still manifests itself in the concepts of "unfairness" and "adverse impact." Furthermore, the determination of the presence or absence of adverse impact on the basis of the impact ratio involves numerous statistical and measurement issues. Use of the impact ratio to establish the prosecutorial posture of an enforcement agency is clearly flawed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

"Inferences from personnel tests and their validity": Correction to Lawshe.

Reports an error in "Inferences from personnel tests and their validity" by C. H. Lawshe (Journal of Applied Psychology, 1985[Feb], Vol 70[1], 237-238). On page 238, line 4, the word "each" appears and should be "such." The sentence will, therefore, refer "to the use of such cognitive processes as inductive and deductive reasoning and such characteristics of temperament as emotional stability and self-esteem." (The following abstract of the original article appeared in record 1985-16032-001.) Contends that despite clear definitions in standard sources, psychologists persistently refer to the validity of tests instead of the validity of inferences from test scores. This persistence leads to references to "kinds of validity" when, in fact, there are "kinds of validity analysis strategies" whereby data are collected or generated to determine or defend the extent, degree, or strength of the inference or inferences that can be made from a set of test scores. It is concluded that content validity analysis strategies are appropriate only when the job behavior under scrutiny falls at the observation end of the continuum; when such behavior approaches the abstract end of the continuum, a construct validity analysis strategy is indicated. (5 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Inferences from personnel tests and their validity.

[Correction Notice: An erratum for this article was reported in Vol 70(3) of Journal of Applied Psychology (see record 2008-10968-001). On page 238, line 4, the word "each" appears and should be "such." The sentence will, therefore, refer "to the use of such cognitive processes as inductive and deductive reasoning and such characteristics of temperament as emotional stability and self-esteem."] Contends that despite clear definitions in standard sources, psychologists persistently refer to the validity of tests instead of the validity of inferences from test scores. This persistence leads to references to "kinds of validity" when, in fact, there are "kinds of validity analysis strategies" whereby data are collected or generated to determine or defend the extent, degree, or strength of the inference or inferences that can be made from a set of test scores. It is concluded that content validity analysis strategies are appropriate only when the job behavior under scrutiny falls at the observation end of the continuum; when such behavior approaches the abstract end of the continuum, a construct validity analysis strategy is indicated. (5 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

An empirical comparison of two methods of test selection and weighting.

Based on data on 9 tests for 2 samples of Navy trainees (N = 184, 176) the Wherry-Doolittle Test Selection Technique was compared with a short-cut method proposed by Jenkins, using a table. The composite r between Grade Point Average and the predicted composite for the Wherry-Doolittle method was .648, while that for Jenkin's method was .646. Jenkin's method takes one-fifth to one-eighth less time than the Wherry-Doolittle. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Obituary: Joseph Tiffin (1905–1989).

Presents an obituary for Joseph Tiffin. Joseph Harold Tiffin (he didn't like the "Harold" and never used the "H") was born in Falls City, Nebraska, on July 4, 1905, grew up in Sioux Falls, South Dakota, and received his bachelor's degree (Phi Beta Kappa) from the University of South Dakota in 1927. As a newlywed, he entered graduate school at the University of Iowa where he was an Eastman Fellow, received both his master's and PhD (1930) degrees under Carl Seashore, and was a fellow of the National Research Council before becoming a member of the University of Iowa psychology faculty in 1931. As Seashore's student and colleague he developed an intense interest in psychophysics and sensory/motor phenomena. (He was the first to photograph the human vocal chords in action.) In the early stages of as-yet-undiagnosed Parkinson's disease, he retired as Professor Emeritus of Industrial Psychology at Purdue University in 1971. Joe and his wife, Mary Edith Straight, moved to Madison, Wisconsin, to be near their daughter and her family. Mary Edith died in 1985, following which Joe was moved to a nursing home. He died March 1, 1989. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Studies in management training evaluation: I. Scaling responses to human relations training cases.

Dimensions of employee-orientation and sensitivity are used in this study of methodology in human relations training. "… is the purpose… to describe a scaling procedure by means of which open-end responses to stimulus films can be reliably scored." After showing 3 films "several groups of academic and industrial Ss" wrote responses to questions relevant to the 2 dimensions of supervisory behavior. 6 sets of responses were scaled by 16 judges, using a forced-distribution scheme. An abbreviated procedure was worked out and scale values obtained "indicate that the master scale method can be utilized with confidence." (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Studies in management training evaluation: II. The effects of exposures to role playing.

(see 33: 11142) How effective is role playing as a tool in management human relations training? "A campus conference group of residential contractor foremen participated in this study." A film was shown and role players enacted completion of the criterion case. There were repeated exposures prepost Employee-orientation comparisons. Impact was the important thing-not repeated exposure. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Of management and measurement.

2 major deficiencies in the field of industrial psychology partially explain the minimal nature of psychology's contribution to the recent development of an increasing quantitative emphasis in American management. "The first of these has been our failure to develop a clear-cut conceptual basis for differentiating between so-called 'basic' and 'applied' research. And second… our tendency in the past to work on problems that fit our methods rather than to devise methods that are appropriate to the major problems of American Management… . As I view the next decade, I see the opportunity for a new era in the field of industrial psychology. Whether it becomes, in fact a new era or not will depend upon the willingness of industrial psychologists to accept the challenge that is currently being presented to them." (PsycINFO Database Record (c) 2010 APA, all rights reserved)