Search

This article presents an analysis of the participation of Texas public school districts in the Advanced Placement (AP) program for academic years ending 2013 through 2019. We develop and estimate to a panel of school district data a hurdle count model composed of: (1) a first stage fixed effect binomial logistic regression model of the probability that a school district opts-in to locally offer AP program courses in a given academic years; and (2) a second stage fixed effect zero truncated Poisson regression model of the number of unique AP courses offered in a given year by opting-in districts. We find that the number of Texas school districts providing local access to AP courses is decreasing through time; however, we also find that districts that do provide local access to AP courses offer increasingly greater numbers of unique AP courses.

This article presents an exploratory case study examining the effects of different early college credit programs on time to baccalaureate degree attainment at a metropolitan Midwestern United States land grant university. We developed a Cox proportional hazards regression model of a students' time to degree attainment as a function of their participation in different early college credit programs while controlling for each student's preenrollment grade point average (GPA), American College Test (ACT) test score, gender, part or full-time enrollment status, ethnicity, and seasonal (Fall, Spring, or Summer) semester of initial matriculation. The most noteworthy finding of our analysis was that each early college credit program appeared to have a positive and statistically significant impact on reducing the time to degree attainment with all other factors being equal. However, the mechanisms through which these programs affect the time to degree attainment appeared to be differentiated by program.

This study compares the performance of longstanding methodological techniques of multinomial logit and ordinal probit models with more recent methods of decision tree and artificial neural network models, and combines individual models into ensembles to test if the amalgamation of the multiple methodologies enhances the classification accuracy of crash injury severity outcomes. The models are estimated using 2002 to 2012 crash data from the Missouri State Highway Patrol and the variables examined include driver characteristics, temporal factors, weather conditions, road characteristics, and injury severity levels. The accuracy and discriminatory power of explaining crash severity outcomes among all methods are compared using classification tables and Area Under the Receiver Operating Characteristic curve values. The Chi-square Automatic Interaction Detection decision tree model is found to have the greatest accuracy and discriminatory power relative to all evaluated approaches. [ABSTRACT FROM AUTHOR]