Development of the Stress Strategy Test as a personnel selection instrument

Abstract: The thesis concerns the development of a computerised test, termed the Stress Strategy Test (SST), intended for use in a personnel selection and placement context. A similar but less extended form of that test, termed the Stroop Strategy Test, developed first, was adjudged to be somewhat less effective as a selection instrument. Both of the tests, concerned with cognition, personality and ability to deal with stress, are based in part on the Stroop test, developed in the first half of the 1930s by J. R. Stroop. That test utilises the fact that it takes longer for persons to name the color in which each of a series of words is printed when the words represent color names but are printed in a color different from the colour the word designates. The greater time such a task takes is termed the ”Stroop effect” or ”interference effect”, The two tests are also based in part on the Serial Colour Word Test (S-CWT), developed by Smith and Neumann in the 1960s. That test utilises the Stroop effect but examines as well changes in performance at such tasks over time. The two present tests, in addition to being computerised, take account of a greater number of variables than are measured in the conventional Stroop test or the S-CWT. The SST is also designed so as to produce what was termed a ”reorientation effect”, which makes the tasks in question more difficult. Three empirical studies were carried out. Studies 1 and 2 concern the Stroop Strategy Test and the Stress Strategy Test, respectively. In both studies, three groups of military men differing in the level of qualification required for the type of jobs to which they had been assigned –primarily on the basis of an intelligence test and an interview – as well as a fourth group of men, found guilty of having committed serious crimes and thus not adjudged to be qualified for military service, were given the test in question. In multivariate analyses carried out in both these studies it was found that the four groups in question could be correctly classified in terms of group membership – on the basis of results of the respective test – with an overall accuracy of 52.6% in the case of the Stroop Strategy Test (consisting of two test series) and with an accuracy of 65.2% in the case of the Stress Strategy Test (consisting of three test series, the third series involving both the Stroop effect and the reorientation effect). In the third study, a large group of military recruits (more than 3000 in all), both men and women being included, were tested with the SST as well as with the two selection instruments currently employed: the intelligence test and the interview. Relations between the SST results and results of the two selection instruments currently employed were investigated, a common variance of 17% and 12% being found in relation to results of the intelligence test and the interview, respectively. The potential usefulness of the SST both in a military context and in other selection contexts is discussed.