Replication Data for: Estimating treatment effects for multiple, continuous treatments: Research success in liberal-arts college economics departments

Data are collected from faculty curricula vitae, institution websites, and where necessary through personal correspondence. Time-dependent variables (like number of years since receipt of PhD) are all clocked at Summer, 2018. We have more data—75 variables in all —but focus here on variables used in this paper’s estimations. Our pre-treatment matching variables include a binary variable for appointment to a research-assistant position while in graduate school, the PhD program’s rank (US News and World Report), the number of peer-reviewed journal articles in the first three years after receipt of the PhD (as an index of research-preparation quality of the PhD program), a binary variable indicating PhDs in traditionally theory-heavy fields (theory, econometrics, international), a binary variable for appointment to a post-doc position at completion of the PhD, a measure of the number of years since receipt of the PhD, and gender (male=0). Our treatment variables record the potential number of sabbatical leaves (given a subject’s time since PhD and institutional sabbatical policy), the number of economist colleagues (excluding self) and number of colleagues in the same subfield of economics, the institution’s average class size, the number of weeks of paid and unpaid family leave, the institution’s annual per-faculty travel funding, the institution’s teaching load (average courses per semester), a binary variable equal to one if the institution offers pre-tenure research leaves, the length of the institution’s tenure clock (in years), a binary variable equal to one if the participant has served as department chair, the institution’s rank among the top 50 national liberal arts colleges (2018 US News, #1 being top-ranked), and a binary variable equal to one if the economics department is housed within a context that is an accredited business program. Our primary outcome variable is the number of articles in journals that appear in the Australian Business Deans Council Journal Quality List.