US FACULTY SALARY AVERAGES: 2021-2022

Author

David J. Bauer, Ph.D.

Published

January 6, 2023

OVERVIEW

Institutions of higher education in the United States report various bits of data to the Integrated Postsecondary Education Data System (IPEDS), including average faculty salaries by rank. The following plot and table provide an accessible means of exploring these data.

You may find these data helpful to inform job-seeking decisions or salary negotiations, or for other purposes that would benefit from comparing average salaries across institutions.

Data are included for academic year 2021-2022, which is the most recent year for which full data are available. Refer to the Appendix for details regarding the data characteristics, data limitations, data sources, variables, renaming and recoding, and filtering.

Briefly, I included each institution’s average salary for full-time, non-medical, instructional staff (i.e., “Faculty”), standardized to a 9-month contract, in the 2021-2022 academic year. Institutions include all public and private (non-profit) 4+ year and 2-4 year institutions (n = 3185).

SALARIES IN 2021

The raincloud plot features a density distribution on top, with a boxplot and individual data points below. Each dot represents an institution’s average salary for faculty of the specified rank, with the median average salary reported above the boxplot.

A data table is also provided. You can sort this by clicking on the column of interest; by default, this is NAME. Click once for ascending sort, twice for descending.

Sort by multiple columns by clicking on the column of primary interest, holding shift, and then clicking one or more subsequent columns of interest. For example, to sort by RANK and then by SALARY AVERAGE: click on RANK, hold shift, and click on SALARY AVERAGE. Note that if you click SALARY AVERAGE a second time (while still holding shift), you can change the order from ascending to descending.

Use the buttons to copy the data (for pasting somewhere else), to save a CSV or Excel file, or to print the table. The print option is useful if you want to save the table as a PDF. Sort the table as desired prior to copying, saving, or printing.

Use the Search box to filter the data by one or more specific observation(s) of interest (e.g., the name of an institution, or a specific academic rank).

CONTACT INFORMATION

David J. Bauer, Ph.D.
Professor of Psychology
Viterbo University
djbauer@viterbo.edu

APPENDIX

DATA CHARACTERISTICS

  • Personnel include full-time, non-medical, instructional staff.
    • These personnel have instruction as their primary job responsibility, but additional responsibilities may include research and/or public service.
    • Administrator and other non-instructional staff salaries are not included. These data are available from IPEDS but not incorporated here.
  • Each institution reports the total number of faculty and the total salary outlay for each academic rank.
    • Total salary outlays have been standardized to a 9-month contract.
    • The average salary by rank is calculated by dividing the total outlay by the number of faculty.

DATA LIMITATIONS

  • The analyses are limited to the salary data provided by institutions to IPEDS.
    • Data for individual faculty are not available. I do not have access to individual salaries, discipline/department, teaching loads, number of course preps, class sizes, number of advisees, committee responsibilities, accreditation requirements, research expectations, etc.
    • I also do not have compensation data for perks and benefits such as health insurance coverage, retirement plan options, tuition remission availability, etc.
    • Salaries are not weighted by any cost of living indices.
    • Salaries are not adjusted for inflation.

DATA SOURCES

All of the data files are publicly available from the IPEDS Data Center. Many additional files, containing many additional variables of potential interest, exist at the same location.

The files are primarily organized by year. For each year of interest, download the following files:

  • “Directory information”. This file contains basic information for every institution (name, location, Carnegie classification, etc.).
    • Example (2020-2021): hd2020.csv (.zip, 1.0 MB).
    • Also download the associated Dictionary file: hd2020.xlsx (.zip, 217 kB).
  • “Number and salary outlays for full-time nonmedical instructional staff, by gender, and academic rank: Academic year ####-##”. This file contains the salary data. When available, use the revised data file, which will be included in the zipped folder.

VARIABLES

The following variables exist in one or both of the two data files indicated above, for each year. I combined these into a single data set for each year by selecting the variables of interest in each file and joining them based on the identification key.

  • UNITID = Unique identification number (data key)
  • INSTM = Institution name
  • CITY = City
  • STABBR - State abbreviation
  • ZIP = ZIP code
  • SECTOR = Institution categorization that combines ICLEVEL with CONTROL.
  • ICLEVEL = Institution categorization based on program year length (4-year, 2-year, <2-year).
  • CONTROL = Institution categorization based on operating officials / funding sources (Public, Private non-for-profit, Private for-profit).
  • C15BASIC = Basic Carnegie classification using criteria updated in 2015.
  • C18BASIC = Basic Carnegie classification using criteria updated in 2018.
  • C21BASIC = Basic Carnegie classification using criteria updated in 2021.
  • ARANK = Academic rank (Professor, Associate professor, etc.).
  • SATOTLT = Number of full-time, non-medical, instructional staff.
  • SAEQ9OT = Total salary outlays of full-time, non-medical, instructional staff equated to a 9-month contract.
  • SAEQ9AT = Average salary for full-time, non-medical, instructional staff equated to a 9-month contract.

Then I created a YEAR variable and joined all of the years together to create a single data set. However, the plot and table above only use data from 2021-2022. Also note that several variables were not displayed in the table, in an effort to reduce clutter.

RENAMING AND RECODING

I renamed and recoded several variables for easier reading and interpretation. Specifically:

  • INSTNM = “NAME”
  • STABBR = “STATE”
  • ICLEVEL = “LEVEL”
    • 1 = “4+ years”
    • 2 = “2-4 years”
    • 3 = “0-2 years”
  • CONTROL
    • 1 = “Public”
    • 2 = “Private”
    • 3 = “For_Profit”
  • C15BASIC = “CARNEGIE_15”
  • C18BASIC = “CARNEGIE_18”
  • C21BASIC = “CARNEGIE_21”
  • ARANK = “RANK”
    • 1 = “Professor”
    • 2 = “Associate”
    • 3 = “Assistant”
    • 4 = “Instructor”
  • SATOTLT = “COUNT”
  • SAEQ9OT = “SALARY_TOTAL”
  • SAEQ9AT = “SALARY_AVERAGE”

I also created the variable CARNEGIE, which basically combined CARNEGIE_15 and CARNEGIE_18 and CARNEGIE_21. Each institution has one of these classifications, depending on year; combining them into a single variable facilitated analyses.

Finally, I converted all ZIP codes to five digits (some were nine).

FILTERING

Several categorical variables include levels that aren’t relevant for comparison purposes; for example, there are 7 levels of Academic rank (ARANK): Professor, Associate professor, Assistant professor, Instructor, Lecturer, No academic rank, and All instructional staff total. Only the first four of these are relevant for my purposes, so I filtered the data to only include those observations.

I also filtered observations based on the following criteria:

  • ICLEVEL: include “4+ years” and “2-4 years” institutions. I removed “0-2 years” institutions.
  • CONTROL: include “Public” and “Private” institutions. I removed “For_Profit” institutions.

Finally, I removed any observations that did not contain complete data.

This initial filtering process leaves a data set of 3185 institutions.