NotesFAQContact Us
Collection
Advanced
Search Tips
ERIC Number: ED449211
Record Type: Non-Journal
Publication Date: 2001-Feb-3
Pages: 23
Abstractor: N/A
Reference Count: N/A
ISBN: N/A
ISSN: N/A
Recent Literature on Whether Statistical Significance Tests Should or Should Not Be Banned.
Deegear, James
This paper summarizes the literature regarding statistical significant testing with an emphasis on recent literature in various discipline and literature exploring why researchers have demonstrably failed to be influenced by the American Psychological Association publication manual's encouragement to report effect sizes. Also considered are defenses of statistical significance. Statistical significance testing can become a measure of the sample size a researcher has used to analyze data, in that the premium placed on obtaining statistical significance overshadows what the data actually indicate about the hypotheses. Perhaps the most productive course of action is to identify a hybrid approach that uses arguments from both sides in the debate over statistical significance testing. Reporting of statistical significance findings may continue, but researchers may be required by scholarly journals to report effect sizes and support for replication and extension studies. (Contains 30 references.) (Author/SLD)
Publication Type: Information Analyses; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A