This is an image that says University of Nevada, Las Vegas

Examining Opposites

When exact opposites divide a construct into two parts, each variable is 0 when the other is positive: They do not correlate -1.  We have created a method of estimating the correlation between the two constructs when each of their measures may not cover all parts of those constructs. This method was described in our 2020 presentation:

Barchard, K. A., & Russell, J. A. (2020, June 1 – September 1).  I’m less and less happy until finally I’m sad: Estimating correlations when variables divide a construct into parts.  Poster presented at the Association for Psychological Science poster showcase, Chicago, IL.      Poster      Handout

To implement this method, use one of the following Excel sheets.

Small file, fast download: Barchard, K. A. (2020, May). Estimating rho(X,Y) when x and y cover arbitrary portions of X and Y (n = 10,000). Unplublished Excel sheet. Available at

http://barchard.faculty.unlv.edu/wp-content/uploads/2020/06/estimating-rhoXY-when-x-and-y-cover-arbitrary-portions-of-X-and-Y-n10000-Barchard-June-6-2020-1.xlsx

More accurate results: Barchard, K. A. (2020, May) Estimating rho(X,Y) when x and y cover arbitrary portions of X and Y (n = 500,000). Unplublished Excel sheet. Available at

http://barchard.faculty.unlv.edu/wp-content/uploads/2020/06/estimating-rhoXY-when-x-and-y-cover-arbitrary-portions-of-X-and-Y-n500000-Barchard-June-6-2020.xlsx