ERIC Number: ED347173
Record Type: RIE
Publication Date: 1992-Jan-21
Reference Count: N/A
Factors That Attenuate the Correlation Coefficient and Its Analogs.
The correlation coefficient is an integral part of many other statistical techniques (analysis of variance, t-tests, etc.), since all analytic methods are actually correlational (G. V. Glass and K. D. Hopkins, 1984). The correlation coefficient is a statistical summary that represents the degree and direction of relationship between two variables. It is important to understand the various factors that affect this coefficient. This paper explains the assumptions underlying the use of the bivariate correlation coefficient, using small heuristic data sets and graphs to make the discussion concrete and accessible to the reader. Attenuation influences on the coefficient of correlation include: (1) departures from linearity; (2) departures from both variables being similarly distributed; (3) using instruments with lower reliability; and (4) using data in which either variable has a restricted range. Six tables and six scatterplots present study data. There is a nine-item list of references. (SLD)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A
Identifiers: Attenuation (Paradox); Linear Relationships; Scattergrams; Variables
Note: Paper presented at the Annual Meeting of the Southwest Educational Research Association (Houston, TX, January 31-February 2, 1992).