College
Buchtel College of Arts and Sciences
Date of Last Revision
2025-01-14 06:54:24
Major
Statistics
Honors Course
STAT498
Number of Credits
3
Degree Name
Bachelor of Science
Date of Expected Graduation
Fall 2024
Abstract
Regression to the mean is a statistical phenomenon that can hide important characteristics of what is truly happening in a research study. Caused by statistical randomness, regression to the mean occurs when extreme values, high or low, are followed by less extreme values. To correctly deal with it, one must understand what it is and how to distinguish its effect on conclusions made from the data. This paper provides examples of regression to the mean in both a medical and academic performance study and explains simple identifiers one can observe. Those are then followed up by the introduction of the usage of control groups and baseline studies to establish a basic understanding of the strength of regression to the mean. Thereafter, a fictitious basketball free throw study is created, and a simple analysis is completed. Afterward, other methods related to regression to the mean are briefly mentioned and readers are directed to additional academic sources.
Research Sponsor
Dr. Richard L. Einsporn
First Reader
Dr. Mark Fridline
Second Reader
Dr. Nao Mimoto
Honors Faculty Advisor
Dr. Nao Mimoto
Proprietary and/or Confidential Information
No
Recommended Citation
Ellis, Hunter, "To Mean or Not to Mean: An Investigation of Regression to the Mean" (2024). Williams Honors College, Honors Research Projects. 1904.
https://ideaexchange.uakron.edu/honors_research_projects/1904
Included in
Applied Statistics Commons, Other Statistics and Probability Commons, Statistical Models Commons
