Contributed Paper C210In session (, )
Using Large Data Sets at the Beginning of Introductory Statistics Courses
AuthorsWilliam Rybolt (Babson College, USA)
George Recck (Babson College, USA)
PresenterWilliam Rybolt (United States)
Instructors often believe that students can better comprehend a statistic by first calculating it using a small data set. When the number of summary statistics such as mean and standard deviation is comparable to the size of the data set, it obscures of the value of summary statistics.
In this presentation, we describe exploratory experiments, which study the value of using a large data set in an introductory probability and statistics course as compared to the more traditional approach. We have chosen a large data set consisting of the amount of electrical power produced by a large array of solar cells. This data set has more than thirty-thousand rows in contrast to typical introductory data sets that often have ten or twenty values.