ERIC Number: ED391990
Record Type: Non-Journal
Publication Date: 1995
Pages: 4
Abstractor: N/A
ISBN: N/A
ISSN: N/A
EISSN: N/A
Available Date: N/A
Inappropriate Statistical Practices in Counseling Research: Three Pointers for Readers of Research Literature. ERIC Digest.
Thompson, Bruce
The research literature provides important guidance to counselors working to keep abreast of the latest thinking regarding best practices and recently developed counseling tools. The purpose of this digest is to highlight a few errors that seem to recur within the literature, and to provide some helpful references that further explore these problems. Insufficient attention to score reliability occurs in between 40 to 50 percent of the published research and is serious because effect sizes and power against Type II error are both attenuated by measurement error. Over-reliance on tests of statistical significance is a major second issue that occurs when researchers incorrectly assume that the p values calculated in statistical significance tests evaluate the probability that results will occur. The null hypothesis and statistical significance is discussed. Finally, this digest examines stepwise methods and asserts that they should not be used. Three problems of stepwise analyses are presented. (JBJ)
Publication Type: ERIC Publications; ERIC Digests in Full Text
Education Level: N/A
Audience: N/A
Language: English
Sponsor: Office of Educational Research and Improvement (ED), Washington, DC.
Authoring Institution: ERIC Clearinghouse on Counseling and Student Services, Greensboro, NC.
Grant or Contract Numbers: N/A
Author Affiliations: N/A