Propensity Score Matching Summary
Matching Parameters
The Matching Parameters table shows the values of the parameters set in the Advanced Models section. See Advanced Models. Match with Replacement is 1 when the corresponding check box is selected and 0 otherwise.
Propensity Score Distribution by Treatment
The Propensity Score Distribution by Treatment plot shows the Matching Statistic on the horizontal axis and the treatment variable on the vertical axis. This plot shows the number of observations falling into certain intervals of the matching statistic. There are three types of points that can appear on the plot:
Matched Obs
(Blue dots) Represents observations that are included in the matched sample.
Outside Support Region/Excluded from Analysis
(Red dots) Represents observations with propensity scores that do not fall between the Propensity Score Minimum and Propensity Score Maximum and are therefore excluded from the matched sample.
Support Region (Unmatched Obs)
(Green dots) Represents observations with propensity scores that fall between the Propensity Score Minimum and Propensity Score Maximum, but are unmatched and are therefore excluded from the matched sample.
Analysis for Matched Data
The Analysis for Matched Data section shows information about the model that is fit on the matched sample. This model is a logistic regression if the outcome is binary and a linear regression if the outcome is continuous. The information in this section is equivalent to the report that is produced by running the Generalized Linear Mixed Modeling (GLMM) personality with default settings, the response variable as the outcome, the treatment variable as the only fixed effect, and the Match ID as a random effect. For more information about all the components in a GLMM report, see “Model Fit Reports”. The information in this report can be used to calculate causal effects based on the matched sample.