IBM Support

Propensity Score Matching in SPSS Statistics

Question & Answer


I'm using SPSS Statistics and need to perform matching of treated cases and untreated controls via propensity score matching. Does SPSS Statistics have a preprogrammed option for such an analysis?


There is no formal procedure within SPSS Statistics for propensity score matching, but two Python-based extensions, FUZZY and PSM, are available from IBM SPSS developerWorks. FUZZY requires at least Version 18 of SPSS, while PSM requires at least Version 1.3.0 of FUZZY and at least Version 20 of SPSS Statistics. Both require an appropriate version of Python to be installed on your computer, followed by installation of Python Essentials, which comes with the installation media. Once Python and the Python Essentials are installed, you can install the FUZZY and PSM extensions. FUZZY.spe is available from and PSM.spe is available from . Once these have been downloaded to your computer, you install them via Utilities>Extension Bundles>Install Extension Bundle. Navigate to the file location, select the file and click Open. Follow the prompts to complete installation of the extension bundle and associated dialog. Do this for both extension bundles, then close out of SPSS Statistics and restart the program. The extensions are available via the Data menu.

[{"Product":{"code":"SSLVMB","label":"IBM SPSS Statistics"},"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Component":"Not Applicable","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"Not Applicable","Edition":"","Line of Business":{"code":"LOB10","label":"Data and AI"}}]

Document Information

Modified date:
16 June 2018