Subset vs. Remainder Contrast (Poster 4)

• In Event: AERA Poster Session Saturday 3:20 p.m.
  In Poster Session: SIG-Educational Statistician Poster Session

Abstract

The purpose of this paper is to derive and illustrate how to contrast a mean or variance of a Subset of subjects with the Remainder of a Total sample using only the Subset and Total sample N, mean, and variance, as the corresponding Remainder statistics will often be unavailable. This has many applications wherein interest is in the difference in a group of interest (Subset) and a larger group (Total) where summary data are available for both, but not the Remainder group. An example comparing the course evaluations of an instructor with those of a department is used to illustrate. A spreadsheet that calculates the appropriate contrasts is offered.

Authors

• Mary G. Lieberman, Florida Atlantic University

• John D. Morris, Florida Atlantic University