Timeline for How exactly does one “control for other variables”?
Current License: CC BY-SA 3.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 23, 2022 at 17:41 | comment | added | AdamO | @mnmn matching is most easily achieved by-design: i.e. participant's recruited based on correlatedness (such as twins) or risk-stratification (such as lung cancer patients to non-lung cancer patients based on smoking status). You can also use it to reduce the size of large datasets, so that the statistics are more appealing to the audience. | |
| Jun 11, 2020 at 14:32 | history | edited | CommunityBot |
Commonmark migration
|
|
| Jan 29, 2020 at 5:46 | comment | added | aghd | IS there any good course/MOOC/book on these topics? | |
| Aug 14, 2019 at 13:05 | comment | added | mnmn | Multivariate models, matching, etc. are all valid techniques, but when does a researcher typically use one technique over another? | |
| May 16, 2018 at 20:20 | comment | added | AdamO | It's an answer to the question that was asked, the good discussion so far is somewhat one-sided in favor of adjustment in multivariate models. | |
| Apr 8, 2018 at 13:27 | comment | added | JackOfAll | Totally over my head. | |
| Apr 6, 2018 at 21:12 | history | edited | AdamO | CC BY-SA 3.0 |
added 1953 characters in body
|
| Apr 6, 2018 at 21:02 | history | answered | AdamO | CC BY-SA 3.0 |