Abstract:

The purpose of this research is to evaluate the effect of computer-assisted (CA) certification standards on student performance in undergraduate statistics classes while controlling for locus of control, gender, age, and three measures of pre-existing ability. Fulton, Mangelsdorff, and Bewley (2009) demonstrated that in-class use of technology normally associated with distance learning improved student outcomes and satisfaction in statistics courses. Both of these works illustrated the efficacy of CA without specifically addressing content certification. Fulton, Ivanitskaya, and Erefeev (2010) confirmed that implementation of more frequent deadlines for content certification correlated with improved performance of graduate students in statistics classes after isolating students’ locus of control. Their study did not evaluate the efficacy of different certification standards (e.g., 70% versus 80% certification standards). This study addresses these shortcomings.
The design for this research is a multi-group pre-post with controls. Hawkes Learning System will provide the CA interventions and certification standards for each major content area (e.g., discrete distributions, continuous distributions, one-sample hypothesis testing, etc.) The Hawkes Learning System provides interactive video and audio coupled with three major learning components: instruct, practice, certify. In addition, the software allows for the manipulation of certification standards. (Hawkes Learning Systems, 2010). The treatment groups follow.
1. Standard pencil and paper homework and in-class instruction, n1=130
2. Hawkes Learning System with certification set at 70%, n2=45
3. Hawkes Learning System with certification set at 80%, n3=45
4. Hawkes Learning System with certification set at 90%, n4=45
For this study, the measure of performance derives from students’ scores on a standardized final examination (ratio-level data). The “ability” construct includes three measures: a pre-test (a standardized algebra diagnostic, ratio-level data), a grade point average calculation (ratio-level data), credit hours completed (ratio-level data). Technological intervention is the manipulated variable and includes software certification of topic areas (rather than the use of traditional homework). Locus of control measurements (a dichotomous internal versus external categorization) derive from the classic survey developed by Rotter (1966). An appropriately transformed dependent variable will allow for Analysis of Covariance (ANCOVA). Results will be available prior to final paper submission.

References:
Fulton, L., Mangelsdorff, D. & Bewley, L. (2009, November). An analysis of statistics education technologies in graduate health and business administration. International Conference of Education, Research and Innovation 2009, Barcelona, Spain. Abstract ISBN: 978-84-613-5536-5, Paper ISBN: 978-84-613-5538-9.
Fulton, L; Ivanitskaya, L., & Erofeev, D. (2010). Frequent deadlines in statistics courses: If health executives learn better when they are enforced, why won't undergraduate students? To appear in the Proceedings of the 2010 Joint Statistics Meeting, Vancouver, BC.
Hawkes Learning System (2010). Hawkes Learning System Statistics Software [software]. Available from http://fileburst.hawkeslearning.com/Setups/STATDIS-Student-Setup-Full.exe.
Rotter, J. (1966). Generalized expectancies for internal versus external control of reinforcements. Psychological Monographs, 80: 1-28.

Keywords:

Computer assisted, experiment, certification, software, statistics.