The effectiveness of redesigning college algebra with a heavy focus on instructional technology

Abstract

Low success rates in college algebra have been an ongoing problem, and mathematics education researchers have continually looked for ways to improve student performance and learning (Aichele, Francisco, Utley, & Wescoatt, 2011; Gordon, 2008; Thompson & McCann, 2010; Twigg, 2011). This study examined the effectiveness of the pilot semester, spring 2012, of a technology-intensive course redesign of college algebra as compared to a traditional lecture approach at a mid-sized, diverse, urban university in the Midwest. Final exam performance was the main measure for assessing student learning outcomes and for testing thirteen hypotheses; DFW rate, the proportion of students withdrawing or earning a grade of D or F, was used for testing one hypothesis. Between the two instruction types, the researcher used a quasi-experimental study to compare overall performance on the final exam, performance on the conceptual and procedural questions of the final exam, performance on the individual questions of the final exam, and DFW rates in the course. Overall final exam performance was also compared within each gender, between genders, within two races/ethnicities, and between races/ethnicities. Additionally, performance on the conceptual and procedural questions was compared within each gender and within two races/ethnicities. Final exams were taken by 170 students, 73 students in the redesign approach and 97 students in the traditional approach. T-tests, analyses of covariance, and two-proportion z-tests were used to investigate the hypotheses. In most hypotheses, there were no statistically significant differences between the two types of instruction. One significant difference was found between African-Americans and Caucasians in both the redesign and traditional sections; however, the covariate, American College Testing (ACT) Mathematics Sub-score accounted for the difference, meaning that type of instruction had no effect. Also, a large difference (p = 0.0026) in favor of students in the redesign occurred on one question of the final exam on which students solved a system of linear equations. Lastly, a large difference in course success was observed in DFW rates in favor of the seasoned, traditional lecture approach. The DFW rate was 41% for the redesign approach, but the DFW rate was only 21% for the traditional lecture approach.