The primary purpose of this study was to determine to what extent the "Rat" (freshman) training program at the Virginia Military Institute alters the anxiety and self-concepts of participants. "Rat" training consists of various activities involving physical and psychological stress, that have been modeled after activities contained in Outward Bound and various military programs. Sixty-eight male first-year cadets at the Virginia Military Institute were randomly assigned to one of five experimental groups. The "Rat" Training Experimental Group (RTE) (N=l3) and the Gas Stressor Experimental Group (GSE) (N=l2) were administered the Tennessee Self-Concept Scale (TSCS) prior to and after participation in the "Rat" training program, while the primary Control Group (C) (N=l4) received only the pre-test and post-test without participating in the "Rat" training program. The RTE group was also administered the State-Trait Anxiety Inventory (STAI) immediately prior to its initial participation in "Rat" training and again following 11 weeks of participation and just prior to a session of "Rat" training activities. The GSE group was administered the STAI prior to participation in a military gas chamber exercise. A post-test of the STAI was administered after 11 weeks participation in "Rat" training and immediately prior to what appeared to be a second exposure to the gas chamber. The C group also received a pre-test and post-test of the STAI, but did not participate in the "Rat" training program. The "Rat" Training Control Group (RTC) (N=l4) participated in the "Rat" training program, but received only post-tests of the TSCS and the STAI. The Gas Stressor Control Group (GSC) (N=lS) was not pre-tested, but participated in "Rat" training and was administered a post-test of the STAI immediately prior to participation in a gas chamber exercise identical to that participated in by the GSE group. A post-test of the TSCS was administered to the GSC group following participation in the program. The "total positive" score from the TSCS and the "state" and "trait" anxiety scores of the RTE, GSE, and C groups were analyzed using a Groups x Tests analysis of variance with repeated measures on the latter factor. In addition, a one-way analysis of variance was performed on the post-test "total positive" TSCS scores and the "state" J and "trait" anxiety scores of all five groups (RTE, RTC, GSE, GSC, C). Where significant differences were found through ANOVA, the Scheffe test was utilized to locate the specific cell mean differences. Comparisons of mean self-concept scores revealed that no significant alterations in self-concept took place as a result of participation in "Rat" training. A reduction in "state" anxiety to the gas chamber exercise did take place, but it was not statistically significant. No such reduction took place with regard to the activities within the "Rat" training program and mean "trait" anxiety scores were not altered significantly as a result of participation in "Rat" training. It was concluded that "Rat" training does not positively alter the personality variables examined. / ED. D.