NotesFAQContact Us
Collection
Advanced
Search Tips
ERIC Number: ED529101
Record Type: Non-Journal
Publication Date: 2012-Feb-1
Pages: 13
Abstractor: As Provided
Reference Count: 14
ISBN: N/A
ISSN: N/A
The Utility of Robust Means in Statistics
Goodwyn, Fara
Online Submission, Paper presented at the Annual Meeting of the Southwest Educational Research Association (35th, New Orleans, LA, Feb 1-4, 2012)
Location estimates calculated from heuristic data were examined using traditional and robust statistical methods. The current paper demonstrates the impact outliers have on the sample mean and proposes robust methods to control for outliers in sample data. Traditional methods fail because they rely on the statistical assumptions of normality and homoscedasticity that are often not met with real data. Robust means are superior due to their ability to maintain power and control for Type I errors. Two robust location estimates, L-estimators (e.g., the trimmed mean and the Winsorized mean) and M-estimators, are reviewed. (Contains 3 tables and 1 figure.)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A