“…What counts as a 'large' number of shared cosponsorships differs for each pair because different legislators sponsor different numbers of bills (i.e., there is substantial variation on the row sums of B), and the bills they sponsor have been sponsored by different numbers of legislators (i.e., there is substantial variation on the column sums of B; see Table 1). These variations are particularly important in an analysis focused on gender because prior research has demonstrated that there are gender differences in the number (Schwindt-Bayer, 2006;Spirou, 2017;Volden et al, 2013;Schmitt & Brant, 2019;Gagliarducci & Paserman, 2016;Anzia & Berry, 2011) and type (Clark & Caro, 2013;Schwindt-Bayer, 2006;Fouirnaies et al, 2019;Swers, 2005;Shim, 2020Shim, , 2021 of bills sponsored. For this reason, we infer legislative collaborations from bill co-sponsorships using the stochastic degree sequence model (SDSM; Neal, 2014), which controls for these variations.…”