Timeline for Correct way to standardize interaction term with one binary and one continuous variable?
Current License: CC BY-SA 4.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 22, 2021 at 1:25 | comment | added | bill999 | I appreciate it. | |
| Feb 21, 2021 at 16:58 | comment | added | EdM |
@bill999 added a bit to the end to address that. Remember that in forming C_S you didn't just center C but you also scaled by the standard deviation. So the coefficients associated with C (both on its own and in the interaction) differ from those associated with C_S, just like coefficients for a distance measured in miles would differ from those for the same distance measured in millimeters.
|
|
| Feb 21, 2021 at 16:55 | history | edited | EdM | CC BY-SA 4.0 |
In response to comment
|
| Feb 21, 2021 at 4:46 | comment | added | bill999 |
Would you also mind including the interpretations of the interaction terms for Model 0 vs. Model 2? This will help me determine which of the two models makes most sense in my application. No worries if not as you have already answered the original question.
|
|
| Feb 21, 2021 at 4:06 | comment | added | bill999 | This helps a ton. Thanks! | |
| Feb 21, 2021 at 1:52 | vote | accept | bill999 | ||
| Feb 20, 2021 at 19:21 | history | answered | EdM | CC BY-SA 4.0 |