Differences on psychological measures related to military attrition

Canada's volunteer military faces a major manning problem. High attrition rates reflect youth indifference to military service, and research indicates that the manpower pool from which recruits are drawn will shrink significantly in the next two decades, exacerbating an already difficult situation.This thesis reports on the first part of a two phase study investigating some of the variables and the dynamics underlying training attrition. In this phase three personality inventories, Holland's Vocational Preference Inventory (VPI), Levenson's Locus of Control Scale (LCS), and Super's Work Values Inventory (WVI) were related to four discrete categories of attrition and to several composite attrition criteria. In addition, Holland's personality-environment congruence hypothesis was tested.Two distinct samples were used: military recruits, and their supervising Non-Commissioned Officers (NCOs). The recruit sample consisted of 795, 17-25 year old, males who successfully completed Canadian Forces Basic Training without being recoursed, as well as 185 recruits grouped into four discrete and three composite attrition categories. The NCO sample consisted of the 30 Squad NCOs in charge of training the recruit sample.Both recruit and NCO samples were administered questionnaires prior to the start of training. The questionnaires, contained the personality inventories indicated, as well as measures of variables to be analyzed in the second phase of this study.Statistical analysis was carried out using (chi)('2), ANOVA, and point biserial corrrelations. Since this was a field study, the level of significance utilized was p < .05.Three of four hypotheses were supported: (a) Recruits who share congruent LCS personality types with their Squad NCO and the modal squad personality type had lower attrition rates, supporting Holland's congruency hypothesis. (b) As a group, recruits with Realistic, Investigative and Conventional personality types, were more congruent with the squad environments, and exhibited significantly less attrition than Artistic, Social and Enterprising types, supporting the congruency hypothesis, and the use of Holland's theory in the analysis of vocational change. (c) A postulated difference in attrition between LCS personality types, Internals, Externals-Chance and External-Powerful Others, was not supported by the data. However, recruits in one of the composite attrition categories scored significantly higher on the External-Chance sub-scale of Levenson's LCS, suggesting poorer adaptation among those who believe that chance has a controlling effect on their lives. (d) Several work values were negatively correlated with the various attrition categories, possibly an outcome of the modal "blue collar" orientation of the sample. However, the "life style" value correlated positively with those who were passing and requested release and the "surroundings" value correlated positively with those who were failing and requested release, suggesting a consistent relationship between certain work values and attrition categories.Overall results confirm the usefulness of utilizing multiple, discrete, and composite attrition criteria and of investigating attrition from a micro-environmental perspective. The study also supports Holland's personality-environment congruence hypothesis, and suggests a consistent relationship between personality variables and attrition