The Prediction of Educational Goals from Selected Variables

This study examines the relationships of five selected predictor variables to a criterion measure, level of educational aspiration. The research hypothesis is that variations in the educational goals of high school seniors correspond to variations in the following socialpsychological factors: sex, scholastic achievement level, scholastic aptitude, socioeconomic status of family, and birth order. Subjects were randomly selected samples of one hundred twelfth graders from each of three demographically different school settings. A multiple regression equation was developed for each sample to determine which combination of predictors and their regression weights most efficiently predicts the criterion index. The three school settings from which research samples were selected are characterized as suburban, middle level in socioeconomic characteristics, and urban or inner city. The three school settings were found to be significantly different in the mean socioeconomic status level. Conventional methods were used to classify sex, grade point average, I.Q. and birth order. The Roe Level of Function system was used to classify occupations of parents and to assign indices of socioeconomic status. Research sub-hypotheses were established and tested by correlational analysis. Results revealed that the suburban setting sample conforms to the highest degree with the research sub-hypotheses. All five independent variables were found to have a significant relationship with the dependent variable, educational goals. The stepwise regression analysis revealed that three variables, scholastic achievement, socioeconomic status, and sex, make significant contributions to the prediction of educational goals. The combination of the three variables and their regression weights accounts for 41.34% of the variance observed in educational goals..