top banner top banner
index
RegularArticles
ReplicationStudies
EditorialBoard
Instructions4Authors
JournalGuidelines
Messages
Submission

Search publications

An SPSS Macro to Compute Confidence Intervals for Pearson’s Correlation

Full text PDF
Bibliographic information: BibTEX format RIS format XML format APA style
Cited references information: BibTEX format APA style
Doi: 10.20982/tqmp.10.1.p029

Weaver, Bruce , Koopman, Ray
29-39
Keywords: Correlation , confidence interval
Tools: SPSS
(no sample data)   (Appendix)

In many disciplines, including psychology, medical research, epidemiology and public health, authors are required, or at least encouraged to report confidence intervals (CIs) along with effect size estimates. Many students and researchers in these areas use IBM-SPSS for statistical analysis. Unfortunately, the CORRELATIONS procedure in SPSS does not provide CIs in the output. Various work-around solutions have been suggested for obtaining CIs for rhowith SPSS, but most of them have been sub-optimal. Since release 18, it has been possible to compute bootstrap CIs, but only if users have the optional bootstrap module. The !rhoCI macro described in this article is accessible to all SPSS users with release 14 or later. It directs output from the CORRELATIONS procedure to another dataset, restructures that dataset to have one row per correlation, computes a CI for each correlation, and displays the results in a single table. Because the macro uses the CORRELATIONS procedure, it allows users to specify a list of two or more variables to include in the correlation matrix, to choose a confidence level, and to select either listwise or pairwise deletion. Thus, it offers substantial improvements over previous solutions to theproblem of how to compute CIs for rho with SPSS.


Pages © TQMP;
Template last modified: 2017-03-12.
Page consulted on .
Be informed of the upcoming issues with RSS feed: RSS icon RSS