Files
Abstract
This study was conducted to assess whether the selected variables of age, gender, grade level, diploma program, rank in class, number of parents in the home, and participation in the free or reduced lunch program predict absenteeism for high school students. The population for this study was students enrolled in a low achieving, low-income high school located in a metropolitan school system. School records for 296 students provided the data for the seven independent variables as well as total absences for the school year. A stepwise multiple regression analysis determined which combination of variables provided the most significant predictors of student absenteeism. None of the variables were statistically significant at the p <. 05 level. However, participation in the free or reduced lunch program and gender did deserve further analysis at .061 and .091 respectively. Regression analyses were performed using participation in the free or reduced lunch program and gender. With a R 2 of .010 for participation in the free or reduced lunch program and a R 2 of .008 for gender, no statistical significance was indicated. Although no significant outliers were present, the large number of absences was questionable. Therefore, a stepwise regression analysis was performed using all variables but only participants with 50 or less absences. At .080 for R 2 and .036 for adjusted R 2 no combination of variables could be considered predictors of absenteeism for students with less than 50 absences. Correlation coefficients were then considered. Gender indicated statistical significant at the p<.05 level with a p value of .005. A regression analysis for gender of students with less than 50 absences was performed. With a R 2 of .024, gender was not statistically significant. A leave-one-out regression analysis indicated that participation in the free or reduced lunch program and gender had the greatest impact on absenteeism.