Our primary research question is that of the comparative eff
Our primary research question is that of the comparative efficacy of accelerated and traditional accounting programs as manifest by success on the CPA exam. We recognize that due to a number of personal and career decisions, not all of those qualified from any of our programs will attempt the CPA exam. We consider three measures of CPA exam performance, where all are conditioned on attempting the exam. We set PASS equal to one for those who reported that they Sennoside C passed the CPA exam and equal to zero for those who reported that they have attempted the exam at any time, but not yet passed all four parts.Table 2 presents the percent of alumni attempting the CPA exam by program. Table 2 also presents the percent of alumni passing the exam by program for those who have attempted the exam. Overall, 66% of those attempting the exam reported that they had passed. We find that 68% and 70% of our summer and weekend accelerated students, respectively, pass the exam, while only 64% of our traditional students have passed. We find a relatively greater percentage of our traditional students attempt the exam and relatively more of our accelerated program students that attempt the exam subsequently pass the exam. We model CPA exam performance as a function of program and control variables. When the dependent variable is binary (pass/not pass), we employ a logistic regression model. When performance is measured as time to pass and the number of attempts required to pass, to control for the impact of outliers we employ robust regression. Prior research in accounting education, particularly the large number of studies devoted to the impact of the 150-h rule on CPA exam pass rates, provides guidance as to the appropriate control variables. Factors incorporated in prior research fall into three broad categories: measures associated with student characteristics and motivation (GPA, college entrance test scores, age, gender), measures of specific preparation and test strategy (advanced degree, completion of 150 h, accounting units, total college units, review courses) and variables related to the institution (AACSB accreditation, faculty research productivity, cost, and selectivity) (Wilson, 2014). College GPA was significant in a number of studies, as were SAT (Scholastic Aptitude Test, a measure of readiness or aptitude for college-level study) and ACT scores upon entrance (Boone et al., 2006, Howell and Heshizer, 2008, Raghunandan et al., 2003, Van Scoyc and Gleason, 1993). We do not have access to student SAT and ACT test scores, but we include their transcript-reported GPA for the accounting courses within our programs (GPAA). Recognizing the commonality of institution, professors, pedagogy, material, and tests among our various programs, we anticipate that GPAA will proxy for differences in a number of otherwise unobservable student characteristics, such as motivation, aptitude, prior exposure, and study habits, though we recognize that if the impact of these characteristics differs when measuring academic versus CPA exam performance, GPAA will be an imperfect summary control. Prior studies of accelerated learning hypothesize that non-traditional students are more mature and motivated than traditional students (Van Scoyc and Gleason, 1993, Charron and Lowe, 2009). We employ age in years at the time of program completion (AGE) as a proxy for the motivational impacts of previous undesirable work experiences, on-the job learning, maturity, the depth of life’s experiences, and the variability of learning styles across age groups (Franklin and Myers, 2016, Trinkle et al., 2016). Many of the firms hiring our graduates provide a declining monetary bonus payment for passing the CPA exam within the first two years after being hired. Additionally, many public accounting firms recruiting from our programs aggressively push their employees to pass the exam and typically make promotion contingent on passing the CPA exam. Consequently, we include the availability of a cash bonus (BONUS) for passing the exam, and the incidence of having worked for a public accounting firm (FIRM) as motivation-related indicator variables. We include possession of an advanced degree (ADVDEG) as an indicator control variable, as this may reflect experience, talent, and motivation (Titard & Russell, 1989). We also use an indicator variable to control for professional certification in another field or area of accounting (CERT), as this can relate to the same factors as for the advanced degree. The directional impact of this variable is not clear – while some alumni may be motivated to obtain additional certifications, others may not see the need for CPA licensure if Langerhans' cells already possess another certification. We include gender (GENDER) as an additional control variable considered in other studies (Charron & Lowe, 2009). Full definitions for each of these variables are found in Appendix A.