Econometric analysis of firm performance and student success

Diem Kieu Nguyen

This dissertation examines factors affecting the performance of Washington log trucking companies and student success at Washington State University (WSU) using econometric models. The first paper investigates the role of ownership and geographic location in the profitability of Washington commercial log trucking companies, using data from a 2007 log trucking survey. Overall, we found that profitability and therefore competitive success of a log trucking firm is significantly influenced by ownership and geographic location. Particularly, ceteris paribus, a sole-proprietorship firm tends to make less profit than a corporation company. Profitability of a firm having close competitors of large sizes is adversely affected. Density positively affects firm profitability, indicating that benefits from clustering outweigh the decrease in hauling rates due to competition of spatially close trucking firms. The second paper applies survival analysis techniques to identify factors affecting success at WSU, regarding timely degree completion and probability of degree attainment. Four models (Accelerated failure time model, Cox-hazard model, Logistic regression, and Multinomial logit model) are estimated, depending upon measure of student success applied. The results suggest that an admission process selecting students based on individual potential for success should consider other factors in addition to high school GPA and SAT/ACT. Furthermore, retention efforts should identify students based on other risk factors, such as financial aid, residency, enrollment status, and running-start credits. The third paper estimates four separate by-spell discrete-time hazard models to identify factors associated with student risks of departures and returns at WSU. Spells are defined as time periods of enrollment or departure. The sample estimated hazard profile indicates that students are at highest risk of leaving before completion during their first year in college. Within each spell, students are more likely to leave and return in initial semesters. Departure is more likely to occur than return. Major findings are consistent with previous retention studies. For instance, student departure is strongly associated with poor high school GPA. College grade performance is the single most important and consistent factor that reduces risk of leaving and increases probability of return. The probability of different events varies across academic programs and among groups of students.

Econometric analysis of firm performance and student success

Files and links (1)

Abstract

Metrics

Details