How to Do a Salary Equity Study: With an Illustrative Example From Higher Education
| Author | Molly I. Beck,Joanna N. Lahey,Lori L. Taylor,Jeffrey E. Froyd |
| Published date | 01 March 2020 |
| DOI | 10.1177/0091026019845119 |
| Date | 01 March 2020 |
| Subject Matter | Articles |
research-article2019
Article
Public Personnel Management
2020, Vol. 49(1) 57 –82
How to Do a Salary Equity
© The Author(s) 2019
Article reuse guidelines:
Study: With an Illustrative
sagepub.com/journals-permissions
https://doi.org/10.1177/0091026019845119
DOI: 10.1177/0091026019845119
journals.sagepub.com/home/ppm
Example From Higher
Education
Lori L. Taylor1 , Joanna N. Lahey1,
Molly I. Beck2, and Jeffrey E. Froyd3
Abstract
Regular salary equity studies can be a best practice among employers committed to
salary equity and fairly managed compensation. While a well-constructed salary study
can identify inequities for amelioration, a poorly constructed study can create rather
than solve problems. Organizations may be deterred from doing these studies because
of their inherent analytical challenges. We provide a guide for human resource managers
describing how to conduct their own salary studies, how to interpret the results, and
how organizations can apply the results. We describe best practices across public sector
organizations and illustrate them with an example from higher education. We also
provide a link to an online appendix containing sample code that can be used to conduct
such analyses using two popular software packages. The twin goals of the article are to
increase the quality of salary analyses while reducing the barriers to conducting them.
Keywords
salary study, practitioner’s guide to salary equity studies, detecting sex, race and
loyalty pay gaps, hedonic wage analysis
In 2016, President Obama asked businesses to sign the White House Equal Pay Pledge.
One of the essential planks of the pledge was a commitment to conducting “an annual
company-wide gender pay analysis across occupations” (“White House Equal pay
1Texas A&M University, College Station, TX, USA
2University of Arkansas, Fayetteville, AR, USA
3The Ohio State University, Columbus, OH, USA
Corresponding Author:
Lori L. Taylor, Department of Public Service and Administration, Texas A&M University, 4220 TAMU,
College Station, TX 77843-4220, USA.
Email: lltaylor@tamu.edu
58
Public Personnel Management 49(1)
pledge,” 2016). At roughly the same time, the U.K. government started implementing
a policy that requires all employers—public and private—with at least 250 employees
to provide annual reports on salary equity (“Equality Act, 2010 (Gender Pay Gap
Information Regulations,” 2017). The implication is clear: regular pay analysis can be
a best practice among employers committed to salary equity.
Unfortunately, there is little evidence that the public sector in the United States—
which could be expected to lead on this issue—has adopted this diagnostic best prac-
tice. A recent study by the U.S. Office of Personnel Management (OPM, 2014) found
that few federal agencies conduct regular salary reviews. Reese and Warner (2012)
found it “extremely difficult” (p. 326) to find anyone at the state level who could say
whether a state was monitoring compliance with laws regarding pay equity. It is
unlikely that local government is doing more.
Fear of litigation may contribute to the public sector’s dearth of regular salary anal-
yses. Indeed, when we began our analyses for a large public-sector organization, the
first stop was with the lawyers. However, carefully conducted salary analyses are as
likely to remove sources of grievance as to reveal them.
A belief that formal salary structures prevent inequity might also inhibit the use of
public sector salary studies. Why invest in analyzing that which by rule cannot exist?
However, the OPM (2014) report makes it clear that they found evidence of inequities
arising from “discretionary authority” (p. 2) even in federal agencies with set salary
schedules.
A more basic reason why the public sector hesitates to conduct regular salary
analyses may be a lack of expertise. A well-executed salary analysis requires tools
and techniques outside the skill set of most traditionally trained human resource
managers, who may be tempted to rely instead on comparisons of average salary by
position or unit when evaluating equity. After all, it seems intuitive that someone
whose salary is close to average for their position is being compensated fairly.
However, comparisons based on average salaries can be misleading. Average sala-
ries can be skewed by the earnings of a small number of individuals, and within-
group comparisons might not be appropriate if there are within-group differences in
worker productivity. It can be equitable for more-skilled managers to earn more than
less-skilled managers, for example.
A well-designed analysis of salary equity accommodates the multidimensional
nature of the organization’s labor force and overcomes many of the drawbacks associ-
ated with average salary comparisons.1 It supports the pay for performance, broad-
banding systems, and monetary bonuses and awards that are part of a manager’s toolkit
to enhance work motivation and reduce turnover among performers (e.g., Lawler,
1990; Rainey, 2014; Thompson, 2007). It can thereby provide a formal way to increase
personnel flexibility in public organizations often seen to be mired in red tape (Feeney
& Rainey, 2010). It also allows administrators to investigate alternative explanations
for patterns that initially appear to indicate bias and to demonstrate, for example, that
average salaries are higher for men than for women not because of bias but because the
men have more professional experience.
Taylor et al.
59
This article draws on the existing literature on salary inequities to provide a detailed
guide for human resource managers looking to conduct their own salary equity analy-
ses. Our twin goals are to increase the frequency with which salary analyses are con-
ducted in the public sector by lowering existing barriers to conducting such analyses,
and to improve the quality of any such analyses conducted. We describe best practices
across a variety of public sector organizations and illustrate the importance of adopt-
ing those best practices using a specific example from higher education. We then pro-
vide the specific code required to conduct such analyses using two widely used
software packages (SAS and Stata), and detailed explanations regarding interpretation
of regression results.
Why Do a Salary Study?
Public organizations may be concerned about two types of salary inequities. The first
kind includes systematic biases in compensation schemes (Herzog, 2008). Previous
salary studies have found systematic bias in wages based on gender (Alkadry & Tower,
2006; Binder, Krause, Chermak, Thacher, & Gilroy, 2010; Ginther, 2004; Ginther &
Hayes, 2003; Rabovsky & Lee, 2017; Sneed, 2007), race and gender (Webber &
González Canché, 2015), or even loyalty to an organization (Masakure, 2016). The
second type of concern is for fairness at the individual rather than group level. Some
individuals may have lower salaries for idiosyncratic reasons unrelated to productivity
or the usual demographic suspects. For example, an individual who had an unusually
productive year during a time when the organization was experiencing a salary freeze
could earn less than another individual whose similarly productive year occurred when
managers were less constrained. This type of inequity—which may have nothing to do
with any systematic bias—may be particularly pernicious when raises are percentage-
based because small differences early in the career of an employee will become mag-
nified over time.
A well-constructed salary study can help an organization determine whether either
type of inequity exists so that ameliorative actions can be taken. In contrast, a simple
comparison of means or a poorly constructed study may not uncover the above ineq-
uities or may incorrectly imply inequities when none exist. This article draws on the
literature from human resource management, economics, and public administration
to provide a practitioner’s guide to conducting a reliable salary review to provide the
state of the art in salary analysis. Throughout, we illustrate our points using a recently
conducted salary analysis for a large public university in the United States.
Salary Studies in the Literature
Typically, salary equity studies use regression techniques to model the relationship
between a measure of compensation, a demographic of interest (such as sex or race),
and an array of other possible explanatory factors, which are referred to as controls.
Controls included in the salary analysis help rule out alternative explanations for the
patterns in the data. (A supplemental table highlighting several examples from the
60
Public Personnel Management 49(1)
literature and describing the controls used therein can be downloaded here: https://
bush.tamu.edu/research/faculty/projects/)
Credible salary studies include controls for differences in worker productivity.
Some studies have used direct measures of productivity such as numbers of publica-
tions (Barbezat, 2004; Binder et al., 2010; Ginther, 2003; Renzulli, Reynolds, Kelly, &
Grant, 2013), grant dollars received (Binder et al., 2010; Renzulli et al., 2013), or
student performance on standardized tests (Meier & Wilkins, 2002). A few have used
worker reports on their own productivity (e.g., Langbein & Lewis, 1998). Most have
used indirect measures of productivity such as the worker’s highest level of education
(e.g., Bidwell, 2011), rank...
To continue reading
Request your trial