Why Propensity Scores Should Not Be Used for Matching
We show that propensity score matching (PSM), an enormously popular method of preprocessing data for causal inference, often accomplishes the opposite of its intended goal - thus increasing imbalance, inefficiency, model dependence, and bias. The weakness of PSM comes from its attempts to approximat...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Cambridge University Press (CUP)
2020
|
| Online Access: | https://hdl.handle.net/1721.1/128459 |
| _version_ | 1826198361564250112 |
|---|---|
| author | King, Gary Nielsen, Richard Alexander |
| author2 | Massachusetts Institute of Technology. Department of Political Science |
| author_facet | Massachusetts Institute of Technology. Department of Political Science King, Gary Nielsen, Richard Alexander |
| author_sort | King, Gary |
| collection | MIT |
| description | We show that propensity score matching (PSM), an enormously popular method of preprocessing data for causal inference, often accomplishes the opposite of its intended goal - thus increasing imbalance, inefficiency, model dependence, and bias. The weakness of PSM comes from its attempts to approximate a completely randomized experiment, rather than, as with other matching methods, a more efficient fully blocked randomized experiment. PSM is thus uniquely blind to the often large portion of imbalance that can be eliminated by approximating full blocking with other matching methods. Moreover, in data balanced enough to approximate complete randomization, either to begin with or after pruning some observations, PSM approximates random matching which, we show, increases imbalance even relative to the original data. Although these results suggest researchers replace PSM with one of the other available matching methods, propensity scores have other productive uses. |
| first_indexed | 2024-09-23T11:03:39Z |
| format | Article |
| id | mit-1721.1/128459 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T11:03:39Z |
| publishDate | 2020 |
| publisher | Cambridge University Press (CUP) |
| record_format | dspace |
| spelling | mit-1721.1/1284592022-10-01T00:53:51Z Why Propensity Scores Should Not Be Used for Matching King, Gary Nielsen, Richard Alexander Massachusetts Institute of Technology. Department of Political Science We show that propensity score matching (PSM), an enormously popular method of preprocessing data for causal inference, often accomplishes the opposite of its intended goal - thus increasing imbalance, inefficiency, model dependence, and bias. The weakness of PSM comes from its attempts to approximate a completely randomized experiment, rather than, as with other matching methods, a more efficient fully blocked randomized experiment. PSM is thus uniquely blind to the often large portion of imbalance that can be eliminated by approximating full blocking with other matching methods. Moreover, in data balanced enough to approximate complete randomization, either to begin with or after pruning some observations, PSM approximates random matching which, we show, increases imbalance even relative to the original data. Although these results suggest researchers replace PSM with one of the other available matching methods, propensity scores have other productive uses. 2020-11-12T19:18:52Z 2020-11-12T19:18:52Z 2019-05 2020-06-12T15:21:06Z Article http://purl.org/eprint/type/JournalArticle 1047-1987 1476-4989 https://hdl.handle.net/1721.1/128459 King, Gary and Richard Nielsen. "Why Propensity Scores Should Not Be Used for Matching." Political Analysis 27, 4 (May 2019): 435-454. © 2019 The Author(s) en http://dx.doi.org/10.1017/pan.2019.11 Political Analysis Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Cambridge University Press (CUP) Other repository |
| spellingShingle | King, Gary Nielsen, Richard Alexander Why Propensity Scores Should Not Be Used for Matching |
| title | Why Propensity Scores Should Not Be Used for Matching |
| title_full | Why Propensity Scores Should Not Be Used for Matching |
| title_fullStr | Why Propensity Scores Should Not Be Used for Matching |
| title_full_unstemmed | Why Propensity Scores Should Not Be Used for Matching |
| title_short | Why Propensity Scores Should Not Be Used for Matching |
| title_sort | why propensity scores should not be used for matching |
| url | https://hdl.handle.net/1721.1/128459 |
| work_keys_str_mv | AT kinggary whypropensityscoresshouldnotbeusedformatching AT nielsenrichardalexander whypropensityscoresshouldnotbeusedformatching |