HIGH SCHOOL FAMILY LIFE EDUCATION: A PROGRAM EVALUATION FOCUSING ON THE MEDIATING EFFECTS OF STUDENT AND PSYCHOSOCIAL VARIABLES ON MARITAL ATTITUDES, SELF-CONCEPT, AND INTERPERSONAL RELATIONSHIPS (AFFECTIVE, CLASSIFICATION, VALUES)

Family life education (FLE) is one of the major primary prevention/mental health curricula available on the elementary and secondary levels. The major affective objectives of FLE include improving self-concept, family/peer relationships, and attitudes toward marriage. Students in seven FLE high school home economic classes (n = 116) were compared to a control group of four non-FLE home economics classes (n = 36), using change scores based on regressed pretest/posttest differences across nine scales measuring self-concept, family and peer relationships, attitude toward marriage, subjective felt gain from course, and average change score. Three psychosocial scales were also given to FLE students (classroom atmosphere, teacher-student relationship, and expectancy of course gain).;Results indicated that FLE students as a whole scored higher than controls only on subjective felt gain scores. Students were then grouped into three levels on the basis of their individual scores on three psychosocial variables. These groups were compared to controls, using a priori one-tailed hypotheses tested by Dunnett's method. Average change score was found to be significant in the predicted direction, as was subjective felt gain, self-concept and willingness to delay marriage. Relationship with mother was found to be marginally significant (p < .10) for female students only. All four remaining comparisons were in the predicted direction.;Statistical tests were also carried out to determine whether two pretest variables, social desirability and family type, were more effective predictors of outcome for FLE groups than controls. Analyses utilizing z tests of the significance of difference between independent rs found social desirability to be more predictive of change scores on four variables (subjective felt gain and relationships with best same-gender friend, best opposite-gender friend, and mother) for the experimental groups versus controls. Being in a non-nuclear family (family type) was a significant negative predictor for four variables (self-concept, delay of marriage, and relationships with father and sibling closest in age) for students in FLE.;Reliability analyses were carried out for the 10 multi-item scales used in the study. Coefficient alphas ranged from .58 to .84, with a median of .77 (M = .75).