The Effects Of Using Visual Statistics Software On Undergraduate Students' Achievement In Statistics And The Role Of Cognitive And Non-Cognitive Factors In Their Achievement

• Main Author: Maxwell, Kori Lloyd Hugh
• Format: Others
• Published: ScholarWorks @ Georgia State University 2014
• Subjects: Statistics Education; Visual Statistics; Cognitive Factors; Non-Cognitive Factors; Statistical Cognition; Achievement Goal Theory
• Online Access: http://scholarworks.gsu.edu/msit_diss/125; http://scholarworks.gsu.edu/cgi/viewcontent.cgi?article=1142&context=msit_diss

This study examined the effects of visual statistics software on undergraduate students’ achievement in elementary statistics and the role of cognitive and non-cognitive factors in their achievement. An experimental design was implemented using ViSta – a visual statistics program. A sample of 273 undergraduate students at a leading, urban, southeastern research university enrolled in six sections of Elementary Statistics were selected and randomly assigned to experimental and comparison groups. The participants completed four surveys, with pre and post-test measures, which assessed their attitudes, statistics self-efficacy, perceptions of their learning environment, and statistical reasoning abilities. To further guide this study, the modified trichotomous framework (Beyth-Marom, Fidler, & Cumming, 2008; Elliot & McGregor, 2001) of goals, cognition, and achievement was used as the theoretical foundation to categorize the cognitive and non-cognitive predictors in relation to student achievement. Two quantitative data analysis methods were utilized. Mann-Whitney tests were employed to determine if there were any statistically significant differences in overall achievement and cognitive and non-cognitive sub-scales between the experimental and comparison groups. Correlation analysis was used to determine if there were any statistically significant associations between the overall grade in the course and the cognitive and non-cognitive sub-scales. For the qualitative data, error analysis was used to determine any underlying processes or misconceptions evident in students’ problem-solving application. Additionally, reliability analysis determined the internal consistency of the data and fidelity of implementation analysis ensured that the intervention was being applied appropriately. In this study, no statistically significant differences in achievement were noted. However, a significant difference was noted in students’ statistics self-efficacy between the comparison and experimental groups. Finally, using the Pearson product moment correlation (r), a statistically significant correlation was found between the overall grade and attitudes towards the course, attitudes towards statistics in the field, interpreting and applying statistical procedures, identifying scales of measurement, and the negotiation scale of students’ learning environment. In order to improve undergraduate statistics instruction, it was recommended that classes should involve more face-to-face engagement with the instructor, focus more on student-centered practices through the use of interactive technology, and incorporate activities from a variety of disciplines.