“…Clustering techniques are widely used in applications such as taxonomy, market research, genetic analysis, and the targeting of medical and health‐service interventions, and have also been used to select comparison groups for purposes of estimating treatment effects with nonexperimental data (e.g., Peck, ; Weitzman, Silver, & Dillman, ). While previous studies on the effect of PFL have used observations from all states other than California (Bartel et al., ; Lichtman‐Sadot & Bell ), or the three next‐largest states, to define the comparison group (Rossin‐Slater, Ruhm, & Waldfogel, ), cluster analysis allows for the creation of a comparison group based on empirical similarity criteria, thus making it less reliant on subjective choices.…”