Tools for Presenting Examples of Sampling Distributions and Hypothesis Testing

• Activity Number: 264
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Section on Teaching of Statistics in the Health Sciences
• Author(s): Mark C. Fulcomer*+ and S. David Kriska and Marcia M. Sass
• Companies: Restat Systems, Inc. and Walden University and UMDNJ
• Keywords: sampling distributions ; hypothesis testing ; power calculations ; significance levels ; one- vs. two-tailed tests ; sample size determination

Abstract

This poster describes three Excel spreadsheets for presenting real-time illustrations of some basic concepts of sampling distributions and hypothesis testing in the classroom. Incorporating "recursive" formulae to simplify calculations (Fulcomer et al., 2007), the first and second spreadsheets (for the binomial and Poisson distributions, respectively) can aid in explaining important properties of sampling distributions for discrete random variables as well as in comparing two competing sets of values for the distributions' parameters. The third spreadsheet utilizes the normal distribution to display power calculations for simple hypothesis testing for a wide variety of instructor-specified scenarios. These can include comparisons of distributions with 641 different means by five levels of sample sizes and four levels of Type I errors (significance), each with both one- vs. two-tailed rejection regions. All three of the spreadsheets have built-in graphs to accompany the tabled calculations. Because of the ease and speed with which examples can be modified, the spreadsheets can also be used to introduce topics related to sample size determination.