Statistics Online Computational Resource

Technology-based instruction represents a new recent pedagogical paradigm that is rooted in the realization that new generations are much more comfortable with, and excited about, new technologies. The free and Internetbased NSF-funded Statistics Online Computational Resource (www.SOCR.ucla.edu) provides a number of educational materials and interactive tools for enhancing instruction in various undergraduate and graduate courses in probability and statistics using observed or computer generated data. SOCR includes class notes, practice activities, statistical calculators, interactive graphical user interfaces, computational and simulation applets, tools for data analysis and visualization. Based on the promising results from our pilot study in 2005-2006, where we saw a consistent trend of improvement in the SOCR treatment group compared to the control group, in terms of quantitative examination measures, our 2006-2007 study involves over 300 UCLA students. We use a cross-over design for one course, (introduction to probability) taught by one instructor, and a randomized controlled study for two different courses (an introductory statistics course for the life sciences and an introductory course in probability). For the cross-over design course, SOCR-based and non SOCR-based activities or homework are assigned and analyzed relative to their performance. In the controlled study, one class uses SOCR the other does not. Several components of the SOCR-based materials are heavily dependent on simulated data to enhance the understanding on distributions, which is a major topic for the classes tested. Little is known about how interactive simulation based applets can enhance students’ learning, understanding and interest in statistics. As a result, separate questionnaires (Felder-Soloman Learning Style inventory, ATS survey, and entry and exit assessment of student knowledge questionnaire) are also collected separately to explore how SOCR affects the performance of students in relation to their learning style, attitudes towards statistics, prior knowledge and demographic characteristics, and reaction to these applets. The first goal of this study is to try to associate the effects of SOCR with students’ individual learning style based on the Felder-Silverman-Soloman index. How these applets assist students with different learning styles is yet to be observed. Our second goal is to assess students’ achievement with a closer look based on homework and activities directly related to SOCR. The students’ feedback and performances from this study tell us how to best optimize the SOCR resources based on their needs. Moreover, its result gives us a more accurate assessment of SOCR specifically and an assessment of internet-based resources in introductory and more advanced courses of statistics and probability.