Timeline for When should one include a variable in a regression despite it not being statistically significant?
Current License: CC BY-SA 3.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 3, 2017 at 16:43 | comment | added | doubletrouble | Feel free to edit. I didn't think he was looking for that kind of depth in the answer, apologies if my brevity led to gross inaccuracy. | |
| Apr 3, 2017 at 15:14 | comment | added | AdamO | This is a very inadequate discussion on the topic of control of confounding. Correlation with the outcome is not a sufficient condition for confounding and can lead to misspecification of causal models by controlling for mediators: This leads to fallacies such as "smoking cessation does not reduce cardiovascular disease risk after controlling for coronary arterial calcium (CAC)". CAC is the primary way smoking gives you heart disease. See Causality by Pearl, 2nd ed, chapter 3 section 3. | |
| Apr 3, 2017 at 14:17 | history | made wiki | Post Made Community Wiki by whuber♦ | ||
| Apr 3, 2017 at 9:58 | comment | added | Maarten Buis | That is not quite true. In order to be a confounding variable it needs to cause the explained variable and the explanatory variable(s) of interest. If the explanatory variables of interest cause the variable, and it influences the outcome, then it is an intervening variable, and you should not control for it (unless you want to decompose the total effect). | |
| S Apr 3, 2017 at 9:51 | history | suggested | Will Vousden | CC BY-SA 3.0 |
Fixes formatting.
|
| Apr 3, 2017 at 9:36 | review | Suggested edits | |||
| S Apr 3, 2017 at 9:51 | |||||
| Apr 2, 2017 at 20:25 | history | answered | doubletrouble | CC BY-SA 3.0 |