Mark Warschauer, University of California, Irvine, School of Education
Di Xu, University of California, Irvine, School of Education
Sarah Eichhorn, University of California, Irvine, Associate Dean for Distance Learning
Mariela Rivas University of California, Irvine, School of Education
Bianca Cung University of California, Irvine, School of Education
NSF EAGER Grant number 1543986, 2015-2017
A strong quantitative background is critical to college study and careers in STEM, yet a large percentage of high school graduates are underprepared for college-level study of mathematics (Biswas, 2007). For that reason, both 2- and 4-year colleges enroll large numbers of students in remedial mathematics courses (Bettinger & Long, 2005). To meet the huge and growing demand for remedial mathematics instruction, colleges are increasingly turning to online instruction (Jaggars & Xu, 2010). Unfortunately, many students struggle with online courses, receiving poorer grades than in face-to-face instruction. This is particularly true for underrepresented minority groups and low-performing students (Xu & Jaggars, 2014). One reason for this may be the lack of opportunities for peer interaction and collaborative problem solving in many online learning environments, both of which have a positive effect on student learning, especially for potentially marginalized students (for overviews and meta-analyses, see Crouch & Mazur, 2001; National Research Council, 2011; Walker & Leary, 2009).
Beginning in Fall 2015, students enrolled in Pre-Calculus will be randomly assigned into four groups: three treatment groups (small group problem solving through face-to-face interaction; small group problem solving through Scribblar; small group problem solving through VirBELA) and one control group (individual problem solving). Replicating the experiment during winter 2016, which typically consists of lower performing students, allows for an analysis of whether the potential impacts of small group problem solving on student learning outcomes may vary under different contexts. Within each treatment group, students will be randomly assigned into small groups that consist of four to five individuals. Group members will be required to solve a total of five problem sets together but turn in answers individually.
Research Aims Guiding the Study:
Aim 1: Examine whether small group problem solving improves student course engagement, performance, and attitude toward STEM and virtual learning in an online pre-calculus class, compared to individual problem solving.
Aim 2: Determine whether the delivery format of small group problem solving (face-to-face, Scribblar, VirBELA) matters in its impact on engagement, performance, and attitude in an online pre-calculus class.
Aim 3: Explore whether these effects vary across different subpopulations of students and different course settings (regular quarters vs. intense summer quarters).
Aim 4: Conduct a rigorous cost-effectiveness analysis to inform policymakers about the potential gains to achievement per dollar spent.