The Place of At-Risk Factors among Students Graduating or Dropping out of High School: A Study of Path Analyses

Abstract

This investigation examined the impact combinations of at-risk factors had on 10,373 high school students in an urban school district from the school years 2000-2001 to 2006-2007. Students were categorized as high school graduates or dropouts. At-risk factors were associated with each student record. Hypothetical structural models were created for Hispanic, White, Black and total students to best estimate if students would drop out of school based on given independent variables. Statistical tests were performed using path analysis, a subset of structural equation modeling (SEM). After performing an analysis on each hypothesized model and testing each for goodness of fit, final models were devised that reflected the data better. The dependent variable for each model was the year a student dropped out of school. The independent variables were the at-risk factors a student had while enrolled in school: being over age, living in poverty, having limited English proficiency (LEP), having low academic achievement, mobility, and suspension/expulsion. Of the students studied, 20.3% dropped out of high school, while 79.7% graduated. Of students who were one year over age, 19.4% of them dropped out. Of students who were two years over age, 62.1% dropped out. The SEMs revealed LEP, an issue primarily impacting this study's Hispanic students, was the strongest risk factor measured of all models. Academic risk factors most impacted Black students while academic risk factors impacted White students the least. This study indicated that different student ethnic groups reacted in varied ways to at-risk variables. This study supported the hypothesis that students represented by multiple at-risk indicators were more likely to drop out of school than those with only one at-risk indicator. Some paths through certain at-risk factors were more dangerous toward graduation than others. The models demonstrated that stronger path coefficients appeared when factors were mediated by achievement rather than when the path was directly linked from the at-risk factors to the year of dropping out. Also achievement was better predicted by standardized test scores from two rather than three years prior to leaving school. The overall results of this research study could help educators address student risk factors more proactively. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone [phone]. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]