Social Preferences Among Male High School Dropouts Enrolled In a Residential Program

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

William G. Wagner

Advisor Department



The purpose of the present study was to evaluate the predictive relationships of male high school dropouts' self-reported aggression, prosociality, self-worth, and perceptions of social acceptance with peer-rated social preferences, or likableness. A mediation model was proposed, and adolescents' overt conduct was expected to function as a mediator between self-reported attitudes and peers' ratings of likability. This is the first known study to evaluate the processes by which social evaluations occur among at-risk male youth with histories of educational, interpersonal, and behavioral maladjustment. The sample consisted of 147 male high school dropouts enrolled in a 22-week military-style residential program located in the Southeastern United States designed to foster positive developmental outcomes and provide the opportunity for GED completion. The Aggression Questionnaire (AQ; Buss & Perry, 1992), Prosocial Tendencies Measure---Revised (PTM---R; Carlo, Hausmann, Christiansen, & Randall, 2003), Self-Perception Profile for Adolescents (SPPA; Harter, 1988), and a peer nomination task were administered during the first 2 weeks and 18th week of the program session. To measure adolescents' behavioral conduct, adult leaders, or cadres, of program-created platoons rated the appropriateness of each participants' behavior at three 6-week intervals using the Conduct Rating Form (CRF). A second measure of conduct, the number of citations for behavioral infractions accrued by each resident during the 18 weeks, was also utilized. Participants' initial scores on the AQ, PTM---R, and SPPA were not found to be significantly related to peer-rated social preference at the end of 18 weeks. Only AQ scores were found to significantly and positively predict participants' inappropriate conduct. Due to the lack of findings between the variables of interest and social preference ratings, the tests of mediation were not conducted. Additional analyses indicated significant positive relationships between initial and final ratings of social preference. A significant and negative correlation between adolescents' conduct and peer-reported social preferences also was found. In other words, male youth disliked during the initial weeks of the session were more likely to be rejected at the end of the program and display more inappropriate conduct. Implications of the current study and suggestions for future researchers are addressed.