2011
DOI: 10.1080/00273171.2011.540475
|View full text |Cite
|
Sign up to set email alerts
|

A Systematic Review of Propensity Score Methods in the Social Sciences

Abstract: The use of propensity scores in psychological and educational research has been steadily increasing in the last 2 to 3 years. However, there are some common misconceptions about the use of different estimation techniques and conditioning choices in the context of propensity score analysis. In addition, reporting practices for propensity score analyses often lack important details that allow other researchers to confidently judge the appropriateness of reported analyses and potentially to replicate published fi… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

1
371
1
9

Year Published

2015
2015
2021
2021

Publication Types

Select...
6
3

Relationship

0
9

Authors

Journals

citations
Cited by 470 publications
(382 citation statements)
references
References 138 publications
1
371
1
9
Order By: Relevance
“…However, when using observational data it is never entirely clear whether observed changes in the outcome can be attributed to the treatment (e.g., a life transition) or whether they have been caused by confounding variables that have not been controlled for. Propensity score matching allows dealing with this problem by adapting observational data in a way that approximates the situation of a randomized controlled experiment (Rosenbaum & Rubin, 1983;Thoemmes & Kim, 2011). More precisely, propensity score matching removes selection effects in the composition of the treatment versus control group ROMANTIC RELATIONSHIPS AND SELF-ESTEEM DEVELOPMENT 13 (e.g., participants who did versus did not experience a life transition).…”
Section: Methodological Problems In Research On Socialization Effectsmentioning
confidence: 99%
“…However, when using observational data it is never entirely clear whether observed changes in the outcome can be attributed to the treatment (e.g., a life transition) or whether they have been caused by confounding variables that have not been controlled for. Propensity score matching allows dealing with this problem by adapting observational data in a way that approximates the situation of a randomized controlled experiment (Rosenbaum & Rubin, 1983;Thoemmes & Kim, 2011). More precisely, propensity score matching removes selection effects in the composition of the treatment versus control group ROMANTIC RELATIONSHIPS AND SELF-ESTEEM DEVELOPMENT 13 (e.g., participants who did versus did not experience a life transition).…”
Section: Methodological Problems In Research On Socialization Effectsmentioning
confidence: 99%
“…In this study, one-to-many matching was used. Oneto-many matching was one of the commonly used matching methods after one-to-one matching (Thoemmes & Kim, 2011) (e.g., Capraro, Capraro, Morgan, Scheurich et al, 2015). The advantage of one-to-many matching was to increase statistical power compared to one-to-one matching (Shadish & Steiner, 2010).…”
Section: Propensity Score Matchingmentioning
confidence: 99%
“…Therefore, in this kind of situation there is a need to assign a control and an intervention group. Propensity score analysis is one method to aid researchers in assigning groups (Shadish & Steiner, 2010;Thoemmes & Kim, 2011). Propensity score matching is the probability of the participant to be assigned to the treatment condition according to the set of observed covariates that are measured before the intervention (Rosenbaum & Rubin, 1983).…”
Section: Propensity Score Matchingmentioning
confidence: 99%
“…Although our sample appears to be small in comparison to large scale surveys that have employed this method, others have shown that matching reduces bias substantially even in small samples (e.g., Gu & Rosenbaum, 1992 use 50 units in each, treatment and control group, see also Rubin, 1979). We employed 1:1 matching using the nearest neighbor method, which is the most often used matching procedure (Thoemmes & Kim, 2011). We used a caliper of .25 as suggested by Rosenbaum and Rubin (1983) to prevent matching of highly dissimilar observations.…”
Section: Analysesmentioning
confidence: 99%