Timeline for How does propensity score matching that uses only a small proportion of eligible patients affect generalizability?
Current License: CC BY-SA 4.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 24, 2020 at 5:24 | comment | added | Noah | @DimitriyV.Masterov I recently reread Rosenbaum & Rubin (1985) and think it might be of some value here. | |
| Jul 26, 2020 at 21:12 | vote | accept | Diana Petitti | ||
| Jul 26, 2020 at 1:01 | comment | added | Noah | Matching allows you to estimate the ATT usually, not the ATE. Using a caliper makes the estimand no longer either the ATT or ATE. An ATE correctly estimated in a sample (e.g., using IPW) generalizes a population from which the sample could be a simple random sample. In your scenario, the ATE would generalize to the population you described. Whether that is better or worse depends on the purpose of the effect estimation. The closer one gets to random sampling from the population of all eligible patients, the broader the implications are for health policy. | |
| Jul 25, 2020 at 11:23 | comment | added | dimitriy | If the target population are the potential patients of the hospital where the study was done, and say the treatment (broadly defined) remains stable, and you have full common support in the PS distributions, you are saying that matching would still not give you the ATE? And if this data contained patients from many hospitals instead of just one, plus the other two conditions, that would not expand the generalizability of the results? | |
| Jul 24, 2020 at 20:32 | comment | added | Noah | No problem. If you feel like I satisfactorily answer your question, please mark the answer as accepted so the question doesn't remain lingering. | |
| Jul 24, 2020 at 17:27 | comment | added | Diana Petitti | Thanks for the comprehensible and comprehensive answer! | |
| Jul 23, 2020 at 7:35 | comment | added | Noah | I recommend Westreich et al. (2019) for a pretty clear discussion. Imai et al (2008) is more technical and broader. | |
| Jul 23, 2020 at 1:50 | comment | added | dimitriy | Do you have a source that you can recommend to understand the last sentence in the first paragraph? | |
| Jul 23, 2020 at 1:20 | history | answered | Noah | CC BY-SA 4.0 |