NotesFAQContact Us
Search Tips
Peer reviewed Peer reviewed
Direct linkDirect link
ERIC Number: EJ870882
Record Type: Journal
Publication Date: 2009
Pages: 12
Abstractor: As Provided
Reference Count: 24
ISSN: ISSN-0950-0790
Replication of Significant Correlations in Small Samples
Lemons, Christopher J.
Evaluation & Research in Education, v22 n2-4 p75-86 2009
Researchers conducting studies involving individuals with exceptionalities are often prevented from involving large numbers of participants in their study samples. When this is the case, some say significant correlations are likely to replicate because the relation between two variables must be robust enough to be detected even with low statistical power. Others say that significant correlations in the context of small sample sizes are not likely to replicate because one or two data points can have "undue influence" over the fit of the data to the regression line. The purpose of this article is to review the relationship between sample size and the replication of significant findings in correlational research. Data simulations are used to illustrate the effect of sample size and correlation magnitude on obtaining statistical significance in randomly drawn samples. Considerations for researchers interested in replicating findings from correlational research are provided. (Contains 2 tables.)
Routledge. Available from: Taylor & Francis, Ltd. 325 Chestnut Street Suite 800, Philadelphia, PA 19106. Tel: 800-354-1420; Fax: 215-625-2940; Web site:
Publication Type: Journal Articles; Reports - Evaluative
Education Level: Kindergarten
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A
Identifiers - Assessments and Surveys: Woodcock Reading Mastery Test