CONDITIONS AFFECTING THE MILITARY UTILIZATION OF PEER RATINGS THE NEWPORT STUDY Ⅲ. FRIENDSHIP CHOICE

摘要

From research completed with 23 trainee sections at the Naval OCS in Newport.nThe following major findings are reported:n1. An average of about. two out of five acknowledged friends were nominated “high” on peer ratings; significantly fewer friends were nominated “high” at later stages of train¬ing than were nominated “high” at the outset of training;n2. In general, significantly fewer friends were nomin¬ated “high” on the forms having to do with “interest in and enthusiasm for the Naval Service”(IE) and “probability of success in 0C3” (OC) than on the forms dealing with “success as a future officer” (FO) or “leadership qualities” (LQ).n3. A “friendship score” based upon the number of friendship choices received by a subject was found to be significantly correlated with peer nomination scores, but not systematically related to the academic performance criterion. Different forms yielded different relationships with friend¬ship—once again IE and OC were significantly lower than FO and LQ.n4. These relationships were not significantly different for the “research” set as against the “administrative” set.n5. When validity coefficients for peer nomination scores were corrected by partialing friendship, it was found that they regularly retained their level of magnitude, thus demon¬strating that the validity of peer ratings—at least, for this criterion—was not unduly altered by a “popularity” factor.