Variables on Scatterplots Look More Highly Correlated When the Scales Are Increased
- 4 June 1982
- journal article
- other
- Published by American Association for the Advancement of Science (AAAS) in Science
- Vol. 216 (4550), 1138-1141
- https://doi.org/10.1126/science.216.4550.1138
Abstract
Judged association between two variables represented on scatterplots increased when the scales on the horizontal and vertical axes were simultaneously increased so that the size of the point cloud within the frame of the plot decreased. Judged association was very different from the correlation coefficient, r, which is the most widely used measure of association.Keywords
This publication has 5 references indexed in Scilit:
- On the Robustness of a Class of Naive EstimatorsApplied Psychological Measurement, 1979
- THE PERCEPTION OF PEARSON PRODUCT MOMENT CORRELATIONS FROM BIVARIATE SCATTERPLOTSPersonnel Psychology, 1979
- Underestimating Correlation from ScatterplotsApplied Psychological Measurement, 1978
- Robust estimation and outlier detection with correlation coefficientsBiometrika, 1975
- The Influence Curve and its Role in Robust EstimationJournal of the American Statistical Association, 1974