Part Correlations
- Entry
- Reader's Guide
- Entries A-Z
- Subject Index
-
A part correlation is a measure of linear association between one variable and another after the variability in at least one other variable is removed from only one of the original variables. The traditional formula for the part correlation between, say, variables X and Y after controlling for Z in Y, denoted rX(Y∙Z), is
where r is the Pearson correlation between two variables.
rX(Y∙Z) is sometimes referred to as a first-order part correlation, to note that the correlation only controls for one other variable. If controlling for two variables, it would be a second-order, and so on. Zero-order correlations do not control for any other variables.
While not obvious from Equation 1, the part correlation is really the correlation between X and the residual of Y after regressing ...
-
-
- [0-9]
- A
- B
- C
- D
- E
- F
- G
- H
- I
- J
- K
- L
- M
- N
- O
- P
- Q
- R
- S
- T
- U
- V
- W
- Z
-
174036- Loading...
Also from SAGE Publishing
- CQ Library American political resources opens in new tab
- Data Planet A universe of data opens in new tab
- Lean Library Increase the visibility of your library opens in new tab
- SAGE Journals World-class research journals opens in new tab
- SAGE Research Methods The ultimate methods library opens in new tab
- SAGE Stats Data on demand opens in new tab