I'd like to make a scatterplot of a partial correlation - so a correlation with covariates. How should I do this in SPSS? If necessary, I could also do it in R. The most important thing is to show the correlation with variables partialled out.
Asked
Active
Viewed 2,913 times
0

kjetil b halvorsen
- 63,378
- 26
- 142
- 467

Janine W
- 1
- 1
- 1
-
Welcome to this site! If this question is only about how to do this is SPSS or R, it is off-topic. Otherwise, please clarify. – chl Jun 20 '14 at 11:15
-
It is - where should I post my question instead? – Janine W Jun 20 '14 at 11:41
-
1Before we migrate this question on Stack Overflow, it would be good if you could include a [minimal reproducible example](http://stackoverflow.com/q/5963269/420055). – chl Jun 20 '14 at 11:43
-
2As [illustrated here](http://stats.stackexchange.com/a/28493/919) and [here](http://stats.stackexchange.com/a/46508/919) (with full `R` code), you can create scatterplots of the *residuals* of variables with respect to the covariates. The correlation coefficients of these residual scatterplots are the partial correlation coefficients. – whuber Jun 20 '14 at 13:52
-
related http://stats.stackexchange.com/q/70322/3277 – ttnphns Jun 20 '14 at 16:50