Introduction: Case studies in test validation

Abstract

Validation studies represent an endangered species. Thirty years ago, it was common to see validation reports in the major journals. In fact, the journal Personnel Psychology had an entire section devoted to validity reports. Today, there are few publicly available validation studies. Personnel Psychology stopped publishing the Validity Informa tion Exchange in the mid 60's. Major research journals stopped accept ing manuscripts describing orthodox validation studies because these studies did not seem to advance the field sufficiently to warrant the space they would take in the journal. The studies were seen as devoted to practice with little enduring value for science. The implication of this disinterest in validation studies is that we know all we need to know about predicting success in jobs. Can this be true? I think not for several reasons. First, the predictors in use have changed substantially. There is less reliance on the general aptitude tests of the 40's and a greater interest in work sample tests, assessment centers, new cognitive ability tests, personality tests, etc. The fact is that the foundation represented by earlier work is a narrow one. Another cause for concern is the nature of work. It has changed substantially and will change still further in the next decade. Machine operator jobs have become machine tending jobs. Secretarial positions have become word processing and data processing positions. The service industry has exploded, creating job titles that did not exist thirty years ago. Still another change is the nature of applicant populations. They are growing more diverse and this trend will continue through at least the next two decades since the aging work population is being replaced with a younger cadre reflecting greater proportions of women, blacks, hispanics, and other traditionally underrepresented groups. In addition,