Reducing Bias and Error in the Correlation Coefficient Due to Nonnormality.
It is more common for educational and psychological data to be nonnormal than to be approximately normal. This tendency may lead to bias and error in point estimates of the Pearson correlation coefficient. In a series of Monte Carlo simulations, the Pearson correlation was examined under conditions...
| Publicado en: | Educational & Psychological Measurement Vol. 75; no. 5; pp. 785 - 805 |
|---|---|
| Autores principales: | , |
| Formato: | Artículo |
| Publicado: |
Sage Publications Inc.
Oct2015
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ssf&AN=109281098&site=ehost-live header: @attributes: shortDbName: ssf uiTerm: 109281098 longDbName: Social Sciences Full Text (H.W. Wilson) uiTag: AN controlInfo: bkinfo: jinfo: jid: 00131644 EPM jtl: Educational & Psychological Measurement issn: 00131644 maglogo: Y pubinfo: dt: Oct2015 vid: 75 iid: 5 pid: 344 pub: Sage Publications Inc. artinfo: ui: 109281098 10.1177/0013164414557639 ppf: 785 ppct: 20 formats: tig: atl: Reducing Bias and Error in the Correlation Coefficient Due to Nonnormality. aug: au: Bishara, Anthony J. Hittner, James B. affil: College of Charleston, Charleston, SC, USA keyword: correlation nonnormal normality Pearson Spearman transformation correlation nonnormal normality Pearson Spearman transformation ab: It is more common for educational and psychological data to be nonnormal than to be approximately normal. This tendency may lead to bias and error in point estimates of the Pearson correlation coefficient. In a series of Monte Carlo simulations, the Pearson correlation was examined under conditions of normal and nonnormal data, and it was compared with its major alternatives, including the Spearman rank-order correlation, the bootstrap estimate, the Box–Cox transformation family, and a general normalizing transformation (i.e., rankit), as well as to various bias adjustments. Nonnormality caused the correlation coefficient to be inflated by up to +.14, particularly when the nonnormality involved heavy-tailed distributions. Traditional bias adjustments worsened this problem, further inflating the estimate. The Spearman and rankit correlations eliminated this inflation and provided conservative estimates. Rankit also minimized random error for most sample sizes, except for the smallest samples (n = 10), where bootstrapping was more effective. Overall, results justify the use of carefully chosen alternatives to the Pearson correlation when normality is violated. pubtype: Academic Journal doctype: Article src: R language: English refInfo: copyright: @attributes: flag: N holdings: @attributes: islocal: N |
|---|