Academic Salary Survey: What’s Your Number?

Patricia Hall and Varghese George, Georgia Regents University, and Ron Wasserstein, ASA

This year’s academic salary survey includes both faculty and nonfaculty statisticians and biostatisticians. We received responses from 58 institutions in the United States. The data included 980 faculty and 80 nonfaculty statisticians, with gender information. The quartiles and 90th percentile are provided in the summaries, whenever possible. All percentiles are rounded to the nearest $100.

Faculty Data

The original academic faculty data set included 980 observations from 58 institutions. Of these, 119 observations were deleted, mostly because of duplications, and a few due to fatal errors. The final analysis included 861 distinct full-time faculty members in 29 statistics departments (N=469), 18 biostatistics departments (N=336), and 10 math sciences departments (N=56) from 53 institutions in the United States.

Table 1 summarizes salary information for full-time academic faculty in statistics departments by rank and years in rank, based on a nine-month salary. Table 2 provides similar information for full-time academic faculty in biostatistics departments, but is based on a 12-month salary. Table 3 summarizes salary information on full-time academic faculty in the math sciences departments by rank, based on a nine-month salary. A few cases of statistics and math sciences faculty with 12-month salaries were adjusted down by a factor of one-fourth, and a few cases of biostatistics faculty with nine-month salaries were adjusted up by a factor of one-third. The quartiles and 90th percentile are provided.

Tables 4, 5, and 6 provide similar percentiles for the groups in Tables 1, 2, and 3, respectively, stratified by gender. There were six observations that did not have gender information and, thus, were not included in the gender analysis.

Nonfaculty Data

Original data included 80 observations from 14 institutions. However, one institution had only part-time nonfaculty and was excluded from analysis. The remaining 76 consisted of 24 at the doctoral level, 50 at the master’s level, and two at the bachelor’s level. The two observations at the bachelor’s level also were excluded because of the small number, resulting in a total 74 observations for the final analysis. Of the 74 individuals, 58 were from biostatistics departments and 16 from statistics departments. Table 7 provides their salary distribution, stratified by highest degree (master’s or doctorate) and years in rank.

Questions regarding this survey report should be addressed to Ron Wasserstein.