The When and Who of Graduation and Dropout

The When and Who of Graduation and Dropout Predictors: A Moderated Hazard Analysis Sara R. Berzenski, Department of Psychology, College of Social and Behavioral Sciences • • • The goals of the present study were: • To examine commonly accessed student success predictive factors for their differential influence across time • To examine the ways in which the effects of some predictors may depend on others, such that risk is magnified for particular subgroups. Hazard analyses were conducting predicting graduation and dropout among students in the College of Social and Behavioral Sciences at CSUN • Separate models for first time freshmen (FTF; N = 1202) and first time transfer students (FTT; N = 1700) • Students who first enrolled between Fall 2007 and Fall 2010, and were followed through Fall 2016. Hazard analysis plots conditional risk across time by examining the proportion of students who experience each “hazard” (i. e. , graduation or dropout) at each time point among those who remain in the data set at that time point (i. e. , have not previously graduated or dropped out). • These dichotomous outcomes then can be predicted by other variables across time points using logistic regression models, which tell us the number of times more likely an outcome is under one condition than another. Statistical interactions between predictors and “time” indicate whether each predictor’s influence varies across time points. Contact sara. berzenski@csun. edu for a copy of the full written report!