Content uploaded by Daniel Ravid
Author content
All content in this area was uploaded by Daniel Ravid on Apr 05, 2022
Content may be subject to copyright.
A META-ANALYSIS OF THE EFFECTS OF EPM
1
A Meta-Analysis of the Effects of Electronic Performance Monitoring on Work Outcomes
Daniel M. Ravid1
Jerod C. White2
David L. Tomczak1
Ahleah F. Miles1
Tara S. Behrend2
1The George Washington University
2Purdue University
This is a post-review, pre-print version of a manuscript accepted for publication at Personnel
Psychology. Please cite as:
Ravid, D. M., White, J. C., Tomczak, D. L., Miles A. F., & Behrend T. S., (2022, in press). A
meta-analysis of the effects of electronic performance monitoring on work outcomes. Personnel
Psychology.
Correspondence concerning this article should be addressed to:
Daniel M. Ravid
Organizational Sciences
The George Washington University,
Washington, DC 20052.
Contact: DRavid@gwu.edu
A META-ANALYSIS OF THE EFFECTS OF EPM
2
Abstract
Electronic performance monitoring (EPM), or the use of technological means to observe, record,
and analyze information that directly or indirectly relates to employee job performance, is a now
ubiquitous work practice. We conducted a comprehensive meta-analysis of the effects of EPM
on workers (K = 94 independent samples, N = 23,461), while taking into account the
characteristics of the monitoring. Results provide no evidence that EPM improves worker
performance. Moreover, findings indicate that the presence of EPM increases worker stress and
strain, regardless of the characteristics of monitoring. Findings also demonstrate that
organizations that monitor more transparently and less invasively can expect more positive
attitudes from workers. Overall, results highlight that even as advances in technology make
possible a variety of ways to monitor workers, organizations must continue to consider the
psychological component of work.
Keywords: electronic performance monitoring, performance management, meta-analysis,
technology,
A META-ANALYSIS OF THE EFFECTS OF EPM
3
A Meta-Analysis of the Effects of Electronic Performance Monitoring on Work Outcomes
As millions of individuals transition to telework amid the COVID-19 pandemic crisis,
organizations must consider the positive and negative consequences of electronic approaches to
training, assessment, supervision, and performance management. Though the pandemic has
dramatically increased the real and perceived need for remote monitoring, extensive electronic
performance monitoring (EPM) practices have already been ubiquitous for many workers. For
example, nurses have been subject to location tracking via GPS (Carr, 2014) and hygiene
tracking via electronic sanitizer dispensers (Levchenko et al., 2011); manufacturing employees
have been asked to wear RFID (radio-frequency identification) technologies to track their
productivity (Ranganathan & Benson, 2020); police-civilian interactions have been captured via
body cameras (Adams & Mastracci, 2018); and Walmart has patented audio surveillance
technology to track employee behaviors as customers check out (Silverstein, 2018).
EPM refers to the use of technological means to observe, record, and analyze information
that directly or indirectly relates to employee job performance (Stanton, 2000). Unlike traditional
close supervision, employers who use EPM can monitor individuals continuously or
intermittently; discreetly or intrusively, and with or without warning or consent (Ajunwa et al.,
2017). Advances in computing technology and the migration of work into cyberspace has
allowed for the monitoring of individuals in a variety of new, invasive, and relatively
inexpensive ways (Holland et al., 2015). As a result, there is increasing development and
proliferation of work monitoring technologies (Golden & Chemi, 2020), with associated risks
that the psychological effects of using such monitoring may be overlooked.
Although the use of technology to observe employee performance is now many decades
old (with early research on the effects of such work practices dating back to the mid-1980s, e.g.,
A META-ANALYSIS OF THE EFFECTS OF EPM
4
Irving et al., 1986), the effects of EPM on work and workers remain largely unclear. Despite
marketing claims that monitoring can be used to protect worker health and safety, workers often
react negatively to these tools. For example, warehouse workers whose idle time is tracked have
reported feeling uncertainty and fear about being punished for not working long enough
(Frenkel, 2021). Other reports show the ease with which monitored workers learn to manipulate
monitoring criteria and “game the system” (e.g., Satariano, 2020). Findings from the academic
literature are difficult to parse. Empirical findings regarding the effects of EPM on workers vary
greatly from study to study (Ravid et al., 2020), and the common treatment of EPM as a unitary
concept (i.e., present or absent) and lack of common language for discussing varying forms of
EPM have made it difficult to evaluate the psychological implications and generalizability of
these findings (Ravid et al., 2020). The need to navigate these contradictions and understand
EPM is critical and timely as proliferation of EPM increases rapidly (Golden & Chemi, 2020).
A recent review article by Ravid and colleagues (2020) proposed a typology of EPM
characteristics as an attempt to organize and explain the effects of EPM, but they stopped short
of empirically testing the proposed framework. As a result, Ravid and colleagues’ review did not
provide any estimates for the effects of varying forms of EPM on work outcomes or offer any
substantive conclusions about the efficacy of such interventions. Ravid and colleagues note
throughout their review the need for further empirical research to help clarify mixed findings
within the EPM literature to help guide EPM in practice.
Our meta-analysis of the EPM literature answers this call, advancing EPM theory and
research in several ways. By applying meta-analysis to this typology, to test the effects of EPM
on work and workers, we estimate the magnitude and variability of relationships that research
finds between EPM and a variety of important work outcomes; we place these effects in context
A META-ANALYSIS OF THE EFFECTS OF EPM
5
with the effects of other common workplace interventions; and we offer substantive conclusions
as to the efficacy of EPM. We include a wide range of outcomes: task performance,
organizational commitment behaviors (OCBs), counterproductive work behaviors (CWBs),
privacy invasion, perceived justice, felt autonomy, work commitment, satisfaction, perceived
support, monitoring acceptance, and stress. In total, we meta-analyze 94 independent samples
comprising 23,461 people—by far the most comprehensive review of the effects of EPM on
work and workers to date. Importantly, we systematically code for the characteristics of
monitoring included in each study, to then meta-analyze how the strength of relationships differ
across varying EPM implementations. EPM is a phenomenon for which practice continues to
outpace research efforts (Ravid et al., 2020). We advance the initial framework proposed by
Ravid et al. (2020) to offer a theoretical and empirical set of findings and corresponding practical
recommendations to guide this quickly evolving domain.
The structure of this paper is as follows: We first review the EPM literature, revealing the
heterogeneity of EPM characteristics studied and methods used to study them. We then review
Ravid and colleagues’ (2020) EPM framework as a means of organizing this heterogeneity.
Because the early EPM literature lacked a common language to characterize EPM, we apply
Ravid et al.’s (2020) framework to form hypotheses and research questions, test them meta-
analytically, and guide the interpretation of meta-analytic results. In conducting this meta-
analysis, we aim to help clarify the effects of EPM with varying characteristics to (1) offer
guidance to researchers and practitioners using and studying EPM; (2) identify areas of the EPM
literature where significant homogeneity/heterogeneity exists; and (3) highlight areas of the EPM
literature that are well studied and those areas that are in need of further research.
Literature Review
A META-ANALYSIS OF THE EFFECTS OF EPM
6
Approaches to studying EPM have varied greatly, resulting in an increasingly disordered
literature that obscures underlying psychological effects of interest. For example, early EPM
research focused on computerized task monitoring (e.g., keystroke tracking on typing speed) as
compared to traditional forms of supervisor monitoring (e,g., Aiello & Svec, 1993). Later
research broadened to include video monitoring implemented for a variety of purposes (i.e.,
observation purposes in Becker & Marique, 2014, evaluation purposes in Claypoole & Szalma,
2019). Still more recent research has focused on the effects of location tracking technologies on
employee performance (e.g., McNall & Stanton, 2011). As specific forms of EPM and EPM
research have broadened over time, researchers have adopted idiosyncratic terminology, further
hindering integrative organization and understanding of EPM. For example, what Bartels and
Nordstrom (2012) term as “administrative monitoring” (monitoring to decide how hard one had
worked in order to justly disperse rewards) is very similar to what Holt and colleagues (2017)
describe as “monitoring for fairness” and to what DeCaro and colleagues (2011) refer to as
“outcome pressure” monitoring.
An additional source of disorder in the EPM literature is the diverse research
methodologies used by researchers, making comparisons across studies challenging. In addition
to experimental work, EPM researchers frequently conduct cross-sectional survey research using
idiosyncratic or self-developed scales (e.g., Arnaud & Chandon, 2013; Jeske & Santuzzi, 2015;
Stanton, 2000). Researchers have also used vignette-based studies to explore the psychological
effects of EPM. Some studies have asked participants to picture themselves in a situation in
which an organization uses various forms of EPM, and then evaluate their reactions to working
in those conditions (Henle et al., 2009; Holt et al., 2017). A common critique of scenario studies
such as these is that they are unrealistic, do not adequately immerse participants in the situation
A META-ANALYSIS OF THE EFFECTS OF EPM
7
of interest, and are not generalizable (see Aguinis & Bradley, 2014 for a review). At present, it is
unclear how well results from these simulation studies correspond to monitored individuals’
actual reactions. Thus, EPM research is highly varied in the monitoring characteristics studied
and the methods used to assess them. EPM researchers often report results in broad terms,
without acknowledging psychologically relevant monitoring characteristics that vary from study
to study. Collectively, these features make it very challenging to detect meaningful patterns in
the EPM literature via qualitative review.
Typology of EPM Characteristics
Ravid and colleagues (2020) proposed a typology of EPM in which monitoring
characteristics – the purpose, invasiveness, synchronicity, and transparency of the monitoring –
interact to affect individual-level work outcomes. Ravid and colleagues’ (2020) typology
provides an organizing framework and common language for discussing and studying EPM
characteristics, and as the most recent and comprehensive framework for EPM characteristics, is
a logical starting point for the current meta-analysis. It is important to note however, that similar
to other classification frameworks (e.g., the Big Five Personality framework) it is likely that
dimensions of this typology can be further parsed to more granular descriptions if desired.
Below, we briefly describe each of the EPM characteristics included in the typology (see Table
A1 in the Appendix for an overview), and offer a rationale for the inclusion of each characteristic
as a determinant of individual-level work outcomes in our meta-analysis.
Purpose
EPM purpose is the communicated function of, or rationale for, EPM use. It has four
categories. Performance Appraisal, Loss Prevention, and Profit EPM (Performance EPM) is
meant to incentivize effort and performance through between-individual comparisons,
A META-ANALYSIS OF THE EFFECTS OF EPM
8
strengthening of performance contingencies, and discouragement of loafing and deviant work
behaviors. For example, monitoring employee computer usage to deter cyber-loafing (Henle et
al., 2009) and using sensors to alert supervisors if stockroom employees are not moving quickly
enough (Yeginsu, 2018) would both be considered Performance EPM. Development, Growth,
and Training EPM (Development EPM) is meant to provide workers with constructive
performance feedback to identify strengths and weaknesses and aid in learning, skill acquisition,
and performance improvement over time. An example would be the use of time tracking
software to provide employees with information they can use to self-monitor and adjust weekly
time usage and self-assess efficiency as desired. Administrative and Safety EPM (Admin/Safety
EPM) is meant to document behavior for legal, administrative, and informational purposes or to
protect employees and organizations from harm. Some examples of Admin/Safety EPM include
video monitoring for job analysis (e.g., to better understand how a job is performed or to
demonstrate how a task is performed for future employees), email monitoring to identify
phishing attempts or potentially dangerous malware, and wearable tracking technologies meant
to identify and warn of employees about environmental hazards. Finally, Surveillance or
Authoritarian EPM (Surveillance EPM) describes monitoring that is implemented without any
explicit rationale, beyond, perhaps, to collect and have access to employee information (Ravid et
al., 2020). In other words, Surveillance EPM describes monitoring without any explicit purpose.
Invasiveness
Invasiveness describes the intrusive and constricting aspect of EPM, especially as it
relates to an individual’s sense of privacy or autonomy. Invasiveness has several sub-elements.
Scope describes the number of ways an individual is monitored (breadth) as well as the degree to
which EPM data are individualized or aggregated at a group level (specificity). The specificity of
A META-ANALYSIS OF THE EFFECTS OF EPM
9
EPM describes the level of analysis at which electronic monitoring takes place, ranging from
most specific (i.e., individual level) to least specific (i.e., organizational level). The target of
EPM refers to the qualitative foci of monitoring including the kinds of information collected.
Informed by early work on surveillance for public welfare (Schoeman, 1979), individual privacy
rights (Schoeman, 1984), and public discourse on the acceptability of differing forms of
employee monitoring (U.S. Congress, Office of Technology Assessment, 1987), Ravid and
colleagues (2020) propose that the most invasive target of EPM is the monitoring of a person’s
personal thoughts, feelings, or physiology (e.g., tracking the content of personal e-mails,
biometric monitoring), followed by targeting a person’s body or physical location (e.g., GPS
tracking, video monitoring), and targeting task performance (e.g., typing speed). Constraints
refer to the extent that an organization limits how and when EPM data can be collected, in
addition to who can access the data and how it may be used (Ravid et al., 2020). For instance, an
organization may enact high levels of constraints on EPM that captures potentially sensitive
material (e.g., physiological data) by specifying exactly when the monitoring will occur and
greatly limiting who has access to the gathered data. Target control refers to the extent that
individuals have control over the methods and timing of monitoring. For example, an individual
who is asked to record themselves performing a task for future employees to use as reference
may have high levels of control to stop, start, pause, delay, and even edit monitoring whereas call
center employees may have little control over when or how their performance is monitored. EPM
with lower levels of constraints and target control are considered more invasive.
Synchronicity and Transparency
The last two characteristics in the Ravid et al., (2020) typology are Synchronicity and
Transparency. Synchronicity describes the temporal characteristics of EPM, both the
A META-ANALYSIS OF THE EFFECTS OF EPM
10
synchronicity of data collection and the synchronicity of delivery of feedback (Ravid et al.,
2020). Highly synchronous data collection involves continuously gathering information about
employee behavior, whereas asynchronous data collection is periodic or passive (e.g., stored for
viewing later). Synchronous feedback systems provide feedback to employees as they work,
whereas asynchronous feedback may be an aggregated report of employee performance. EPM
Transparency describes the degree to which individuals have access to information regarding
monitoring characteristics (Ravid et al., 2020). Transparency is a continuous measure ranging
from the provision of no information about EPM practices, to the provision of all details (e.g.,
purpose, invasiveness, synchronicity).
Hypotheses and Research Questions
Given the complexity of EPM and the variability in methods used to study EPM, the
current study aims to test a variety of hypotheses and research questions using meta-analysis. An
aim of the current paper is to help clarify the effects of differing EPM characteristics across the
broad range of work relevant outcomes that are most represented in the EPM literature at the
current time. We group these work outcomes into the broad categories of performance, attitudes,
and stress/strain, and offer hypotheses and research questions for each. A great deal of variance
is certain to exist within each outcome variable; we explore these differences by testing study-
level moderators. We organize our hypotheses using Ravid and colleagues’ (2020)
aforementioned typology of EPM characteristics. Given the great deal of variability in design
and use of EPM, we chose not to formulate hypotheses as to any main effects of EPM and
instead offer hypotheses regarding the moderating effects of EPM characteristics. For each
characteristic in the typology (i.e., purpose, invasiveness, synchronicity, and transparency), we
discuss relevant scholarship pertaining to the moderating effects of the characteristic on the
A META-ANALYSIS OF THE EFFECTS OF EPM
11
relationship between EPM and performance, attitudes, and stress/strain. Where theory and
evidence regarding the effects of an EPM characteristic on a work outcome is insufficiently
developed, we propose general moderation hypotheses or research questions.
Purpose as a Moderator of the Effects of EPM on Work Outcomes
EPM Purpose and Performance
Purpose is likely to moderate the effect of EPM on performance. Social information
processing theory (Salancik & Pfeffer, 1978) proceeds from the assumption that individuals use
the activities and words of others to infer the social acceptability of their own behaviors and to
understand behavior-outcome linkages. Individuals are likely to interpret the communicated
purpose of monitoring as a signal about what behaviors an organization values and rewards, and
therefore where effort and attention should be placed (Stanton & Julian, 2002). For instance,
individuals monitored for performance purposes may focus narrowly on task performance;
individuals monitored administrative purposes may attend to administrative compliance;
individuals monitored for surveillance purposes may refrain from engaging in observable CWBs;
and individuals monitored for development purposes may focus on skill acquisition and mastery
believing these behaviors will be valued by the organization. Differences in where individuals
place their attention and effort at work are likely to lead to differences in performance
(Northcraft et al., 2011).
Further, EPM systems with different purposes are likely to send different messages about
an individual’s value and standing, creating differences in motivation to perform. McNall and
Roch (2007) found, in support of this notion, that EPM purpose influenced performance through
its impact on interpersonal justice perceptions and trust in management. Wells et al., (2007)
found that employees were more motivated to help their organization achieve goals when
A META-ANALYSIS OF THE EFFECTS OF EPM
12
monitoring was for development rather than for performance appraisal purposes; and Jeske and
Santuzzi (2015) found that monitoring with different purposes (i.e., for worker safety, to deter
resource abuse) differentially influenced workers’ willingness to engage in OCBs. Thus,
different purposes for monitoring are likely to have different effects on individuals’ motivation
to perform, ultimately leading to differences in performance.
Despite evidence for purpose as a moderator of the effects of EPM on performance, the
current state of EPM scholarship makes it difficult to predict the precise form/direction that the
moderation will take, particularly when examining performance as defined broadly. Evidence
suggests that as EPM focuses attention towards one aspect of performance, it may divert
attention from other aspects of performance (Larson & Callahan, 1990; Stanton & Julian, 2002),
but it is unclear if this is true in all cases. For instance, no research has examined whether
focusing one’s attention on administrative or safety compliance through Admin/Safety
monitoring affects other aspects of performance (e.g., task performance, OCBs), making it
difficult to predict the precise form of this moderating effect. Also unclear are the overall
performance effects of Surveillance EPM, which may direct employee effort towards work tasks
while also communicating to individuals that they are not trusted, perhaps decreasing motivation
to perform. Thus, instead of specific hypotheses about the form/directionality of the moderating
effects of each category of purpose, we pose the following general hypothesis:
Hypothesis 1a: The communicated purpose of EPM will moderate the effects of EPM on
performance.
EPM Purpose and Attitudes
Purpose is also likely to moderate the effects of EPM on attitudes. Nishii and colleagues’
(2008) HR practice attribution model argues that people attach causal attributions (i.e., an
A META-ANALYSIS OF THE EFFECTS OF EPM
13
underlying purpose) to HR endeavors, and these causal attributions influence attitudes at work.
Nishii and colleagues proposed that individuals respond positively to HR practices perceived as
for their benefit and negatively to HR practices perceived as for organizational benefits (e.g.,
worker exploitation, cost cutting). Similarly, Hovorka-Mead and colleagues (2002) drew on
sociological and communications theory (e.g., Benoit, 1995; Scott & Lyman, 1968) to suggest
that individuals use the communicated purpose of EPM to evaluate the extent that monitoring is
justified; when individual’s perceive that monitoring is less justified, they are more likely to
respond with negative attitudes.
Again, it is difficult to predict the precise form/directionality by which purpose
moderates the effects of EPM on attitudes. Research is still needed that explores the attributions
employees might assign to each category of EPM purpose. Employees may perceive
Performance EPM, for example, as implemented for organizational rather than employee benefit
due to the evaluative nature of the monitoring; or they may perceive Performance EPM as for
employee benefit due to EPM’s ability to distribute rewards and punishments more fairly.
Similarly, Admin/Safety EPM may be perceived as implemented for employee safety and well-
being or may be perceived as paternalistic or implemented as a means of protecting the
organization from employee legal claims. Although individuals are likely to view differing EPM
purposes as more or less justified affecting their attitudinal responses to EPM, the precise
magnitude and directionality of these differences are still unclear. We therefore offer the
following hypothesis:
Hypothesis 1b: The communicated purpose of EPM will moderate the effects of EPM on
attitudes.
EPM Purpose and Stress/Strain
A META-ANALYSIS OF THE EFFECTS OF EPM
14
Models of evaluation anxiety (e.g., Zeidner & Matthews, 2005) and role theory (Katz &
Kahn, 1978) support the notion of purpose as a moderator of the effects of EPM on stress/strain
but suggest different forms that this moderation may take. The communicated purpose for EPM
is likely to signal to employees the degree to which they are working in an evaluative context,
influencing their levels of evaluation apprehension and anxiety. Nebeker and Tatum (1993)
proposed that stress might only result when EPM is paired with goals and performance-
contingent rewards, suggesting that EPM for non-evaluative purposes (e.g., Development,
Admin/Safety) may produce relatively little stress. However, different purposes for monitoring
are also likely to vary in the degree to which they create uncertainty regarding role expectations,
and thus role related stress (Katz & Kahn, 1978; Sonnentag & Frese, 2013). Highly evaluative
EPM may help clarify organizational expectations through incentives and punishments attached
to monitoring, mitigating uncertainty and the resulting stress. Sherif and colleagues’ (2021)
qualitative study supports this notion; for example, a participant in that study reported that the
evaluative EPM system in their workplace provided them with clarity about where to focus their
attention and how to spend their time. Monitoring for other purposes such as Admin/Safety or
Surveillance may provide little role clarity to those being monitored. Thus, it is not clear the
precise manner/form by which purpose is likely to moderate the effect of EPM on stress/strain.
We therefore pose the following general hypothesis regarding the moderating effect of purpose:
Hypothesis 1c: The communicated purpose of EPM will moderate the effects of EPM on
stress/strain.
Invasiveness as a Moderator of the Effects of EPM on Work Outcomes
EPM Invasiveness and Performance
Several theoretical frameworks suggest that invasiveness is likely to moderate the effects
A META-ANALYSIS OF THE EFFECTS OF EPM
15
of EPM on performance such that more invasive monitoring has a more positive effect on
performance. Goal-setting theory (Locke & Latham, 2006) and self-regulation theories, for
example, suggest the importance of specific behavioral feedback for goal-achievement and work
performance (Kanfer et al., 2017). More invasive EPM (e.g., monitoring with greater breadth,
greater specificity) is likely to provide individuals and their organizations with more detailed and
specific performance-related information, supporting performance improvements. Individuals
monitored by a low-specificity EPM system (e.g., group monitoring) may feel more tempted to
engage in social loafing than those experiencing EPM at higher specificity, and they may find the
feedback from monitoring too general to improve their individual performance. Consistent with
this idea, Earley (1988) found that computer-based performance feedback of magazine
subscription employees enhanced performance when the feedback received from the system was
specific but not when it was general.
Agency theory (Jensen & Meckling, 1976) provides further theoretical justification for
the positive effects of invasiveness on performance. Agency theory assumes that individuals
know more about their performance behaviors at work than others do and can behave
opportunistically and in self-interested ways due to this information imbalance. EPM provides
organizations with access to employee information, thereby reducing the information imbalance
between individual and organization and limiting opportunities for individuals to engage in
counterproductive behaviors (Alge & Hansen, 2014). According to agency theory, the
performance benefits of EPM should increase as EPM systems transfer more informational
control from employee to organization. Thus, we pose the following hypothesis:
Hypothesis 2a: Invasiveness will moderate the effects of EPM on performance such that
more invasive EPM will be associated with more positive performance.
A META-ANALYSIS OF THE EFFECTS OF EPM
16
EPM Invasiveness and Attitudes
We expect that invasiveness moderates the effects of EPM on work attitudes such that
more invasive EPM is associated with more negative attitudes. More invasive EPM (e.g., broader
scope, more personal target, less target control) is more likely to be perceived as a threat to
individual behavioral freedoms. Psychological reactance theory (Brehm & Brehm, 1981)
suggests that individuals perceive having certain behavioral freedoms that, when threatened or
eliminated, produce an aversive motivational state. This motivational state, state reactance, is
directed towards reasserting behavioral freedom and is associated with negative cognitions and
affect (Rains, 2013). Results from a recent study by Yost and colleagues (2019) suggest that
organizational monitoring policies may indeed produce state reactance in employees; the study
also demonstrated that greater state reactance was associated with negative attitudinal work
outcomes (increased anger and negative cognitions; Yost et al., 2019). Second, more invasive
EPM is likely to capture more non work-related information than less invasive EPM, as the
breadth of behaviors captured is wider, and it may target more personal data. The work-
relatedness of personal information captured by monitoring may moderate the relationship
between EPM use and perceptions of privacy invasion, such that the relationship is lessened
when monitored content is perceived as work-related (Alder & Tompkins, 1997); thus we expect
individuals to respond more negatively more invasive EPM.
Hypothesis 2b: Invasiveness will moderate the effects of EPM on attitudes such that more
invasive EPM will be associated with more negative attitudes.
EPM Invasiveness and Stress/Strain
We expect more invasive EPM to be associated with greater levels of stress/strain, for
several reasons. First, Karasek’s (1979) job demands-control model argues that autonomy in the
A META-ANALYSIS OF THE EFFECTS OF EPM
17
form of decision-making authority instills a sense of control in workers, thus reducing their
feelings of stress or strain in otherwise demanding jobs. By definition, an EPM system that
provides workers with authority over monitoring decisions is less invasive than one that does not
provide such affordances. Second, highly invasive EPM may create a sense of privacy invasion
among workers, and privacy-invading technologies are often associated with increases in worker
stress (Tarafdar et al., 2007). Thus, high levels of EPM invasiveness should be associated with
stress reactions from workers.
Hypothesis 2c: Invasiveness will moderate the effects of EPM on stress/strain such that
more invasive EPM will be associated with greater levels of stress/strain.
Synchronicity as a Moderator of the Effects of EPM on Work Outcomes
Relative to purpose and invasiveness, scholarship regarding the moderating effects of
synchronicity is far less developed. Few studies have directly examined the effects of
synchronous versus asynchronous monitoring and monitoring feedback on work outcomes.
Although highly synchronous EPM may be perceived as more restrictive than intermittent or
delayed data collection and storage, research suggests that individuals may prefer the
predictability of continuous collection to the unpredictability of intermittent monitoring (Jeske &
Santuzzi, 2015). Compared to intermittent monitoring, however, individuals may perceive more
continuous monitoring as more thoroughly and accurately capturing typical performance within
and across individuals; and, therefore, it may be perceived as more procedurally fair (McNall &
Roch, 2007)
In general, timely feedback is considered useful for learning and skill development
(Northcraft et al., 2011), suggesting that individual performance may benefit from greater
synchronicity in EPM feedback. However, it is also possible that as compared to asynchronous
A META-ANALYSIS OF THE EFFECTS OF EPM
18
feedback (e.g., a summarized report of aggregated performance behaviors) synchronous
feedback delivery may increase evaluative apprehension (i.e., anxiety over negative evaluation)
due to more detailed real time behavioral feedback, which in turn may inhibit feedback
integration and learning (Green, 1983). Due to a lack of research and scholarship addressing the
effects of EPM synchronicity on individual-level outcomes, we pose general research questions
regarding the potential moderating effects of synchronicity.
Research Question 1: Does the synchronicity of EPM moderate the effect of EPM on (a)
performance, (b) attitudes, or (c) stress/strain?
Transparency as a Moderator of the Effects of EPM on Work Outcomes
Theory regarding the effect of transparency on individual level outcomes is well-
developed and clear, suggesting that greater transparency in monitoring is likely to produce more
positive work outcomes. Greater transparency about EPM should help clarify the goals and
purpose of EPM, and thus guide individuals towards organizationally desired outcomes and more
positive performance (Locke & Latham, 2006). Further, theories of workplace justice (e.g.,
Greenberg, 1987) argue that organizations that are honest and transparent about organizational
practices will be perceived as fairer than organizations that are not. Greater perceptions of
justice, in turn, have been shown to relate to a variety of positive work attitudes and behaviors
(e.g., satisfaction, commitment, intention to stay, leader member exchange; Cohen-Charash &
Spector, 2001). Indeed, Posthuma and colleagues (2018) have discussed the concept of
transparency in a broader HRM context, theorizing that rich, clear communication between an
organization and its workers will result in more favorable performance and attitude outcomes.
Additionally, greater transparency about the nature of monitoring should help diminish
uncertainty about the monitoring and therefore reduce stress/strain associated with uncertainty.
A META-ANALYSIS OF THE EFFECTS OF EPM
19
Therefore, we propose the following hypotheses:
Hypothesis 3: Transparency will moderate the effects of EPM on performance such that
more transparent EPM will be associated with (a) more positive performance, (b) more positive
attitudes, and (c) lower levels of stress/strain.
Method
We conducted two sets of analyses based on the two different approaches that researchers
have used when studying EPM. The first set of analyses (Presence analysis) includes primary
studies that compared electronically monitored and unmonitored individuals, i.e., presence
versus absence of EPM. These studies could be experimental (e.g., an experiment with an EPM
condition and a non-monitored condition) or non-experimental (e.g., a survey study in which
some respondents indicated the presence of EPM at their work and others indicated that EPM
was not present at their work). In the Presence analysis, we examine the overall effects of the
presence/absence of EPM on work outcomes and examine EPM characteristics as between-study
moderators of these main effects. The Presence analysis allows us to answer the question, “what
is the effect of EPM with a specific characteristic as compared to no EPM at all?”
The second set of analyses (Degree analysis) includes primary studies in which the
presence of monitoring was a constant (i.e., no non-monitored individuals). These studies
measured or manipulated monitoring characteristics at the within-study level. For instance, an
experiment in which one group of participants is monitored more invasively than another group
of participants or a non-experimental study in which respondents rate on a scale how transparent
their organizations are about EPM practices would both be included in the Degree analysis. In
these cases, primary studies were typically more explicit about the monitoring characteristics that
were measured or manipulated. In the Degree analysis, we examine the relative effects of
A META-ANALYSIS OF THE EFFECTS OF EPM
20
varying levels of an EPM characteristic (e.g., more or less transparency) on work outcomes and
answer the question, “what is the effect more or less of a particular EPM characteristic?” We
were able to test any given hypothesis or research question with either the Degree or EPM
Presence analysis, or both, based on the body of primary studies available.
Inclusion Criteria
We set eight inclusion/exclusion criteria for primary studies. First, to qualify for
inclusion, articles must have focused on EPM specifically. Articles that only examined
traditional or broad performance monitoring rather than EPM were excluded. Second, articles
must have examined EPM in a work-related context. We considered experiments in which
participants were instructed to engage in a work task (e.g., data entry, memory task, card sorting)
or training exercise (e.g., learning a skill in Microsoft Excel) to be work-related. We excluded
articles that examined EPM in non-work settings such as consumer habit monitoring.
Third, articles must have reported Pearson correlations or descriptive statistics that could
be transformed to a Pearson correlation for the relationship between an EPM characteristic and a
work-related outcome. Articles that did not provide sufficient information to calculate a Pearson
correlation, and whose authors did not or could not provide the needed information upon e-mail
request, were excluded. Fourth, articles must have been conducted at the between-person level.
Fifth, we excluded studies in which authors clearly used the same dataset and reported the same
correlations in more than one published study, unless different outcomes were considered in both
studies. In cases in which both theses/dissertations and published versions of these
theses/dissertations were obtained, only the published work was included
1
.
1
In the case of Zweig and Webster’s (2002) Study 1 and Zweig’s (2001) doctoral dissertation, the relationships
reported overlapped completely. However, the doctoral dissertation reported descriptive statistics that were
unreported in the published study. We used descriptive statistics from the dissertation to supplement coding for the
published study.
A META-ANALYSIS OF THE EFFECTS OF EPM
21
Sixth, when an article reported results obtained from multiple independent samples, each
sample was included separately (e.g., Claypoole & Szalma, 2019; DeCaro et al., 2011). Seventh,
because our primary interest is in the effects of EPM on individuals who are monitored, articles
in which individuals were asked for their perception of EPM use on others (e.g., Kaupins &
Coco, 2017; in which HR managers rated the ethicality of using various monitoring techniques
on other employees) or in which EPM was included as a dependent variable (e.g., Alge et al.,
2004) were excluded. We also excluded articles in which EPM was operationalized using an
attitudinal measure (e.g., Holman et al., 2002; in which respondents rated the amount that they
were electronically monitored at work on a scale from “too little” to “too much”). Finally, for the
EPM Presence analysis, to qualify for inclusion, articles must have included some individuals
who were not electronically monitored to serve as a comparison in the calculation of
correlations. To be included in the EPM Degree analysis, articles must have included EPM that
varied on at least one psychological characteristic (e.g., EPM with more control as compared to
EPM with less control; McNall & Stanton, 2011).
Literature Search
We conducted an extensive literature search between June and October of 2019 with
supplementary searches conducted in December of 2019, August of 2020, and February of 2021.
The literature search for the EPM Presence and EPM Degree analyses was conducted
concurrently and studies were coded as Presence or Degree later. We started by searching the
Web of Science electronic database using the following advanced search:
“electronic monitoring” AND “performance”) OR TS= (“electronic performance
monitoring”) OR TS = (“computer monitoring” AND “performance”) OR TS =
(“computer surveillance” AND “performance”) OR TS = (“employee monitoring”) OR
A META-ANALYSIS OF THE EFFECTS OF EPM
22
TS = (“work monitoring”) OR TS = (“performance monitoring”) OR TS =
(“performance feedback” AND (“computer” OR “electronic”)) OR TS = (“computer-
aided monitoring” AND employee) OR TS = (“monitoring and performance” AND
“computer”)
This search yielded 5664 results. We next refined results to only include articles from
psychology-, management-, business-, and human-resource-related fields. The refined search
yielded 941 articles. We next read through titles and abstracts of the articles collecting citations
to those articles that were relevant. This process resulted in citations to 102 articles.
In the next step, we read through each article to select only those empirical articles that fit
our a-priori inclusion criteria. We conducted forward searches and examined the reference lists
of all retrieved articles to collect any additional articles not found in our initial search. We also
searched the reference lists of several reviews of employee monitoring (Alge & Hansen, 2014;
Jeske & Kapasi, 2017; Stanton, 2000) including a recently published review of the effects of
body-worn cameras (BWC) on police officers (Maskaly et al., 2017) and conducted follow up
searches using Google Scholar, PsycINFO, and JSTOR to identify additional articles that fit the
inclusion criteria.
Consistent with meta-analytic best practices (Aguinis et al., 2011) we sought to actively
include unpublished master’s theses, doctoral dissertations, and conference papers in our meta-
analysis. We conducted a search in ProQuest Dissertations and Theses using the same
combination of keywords as those above and collected citations for articles that appeared
relevant. We then read through each dissertation selecting those that fit our inclusion criteria. We
conducted additional searches within the conference programs of the Academy of Management
Annual Meeting, the Society for Industrial-Organizational Psychology Annual Conference, and
A META-ANALYSIS OF THE EFFECTS OF EPM
23
the American Sociological Association Annual Meeting from 2010 to 2019 and contacted
authors whose abstracts focused on electronic performance monitoring. We also reached out to
several authors who had published EPM-related research papers previously to inquire about
unpublished and in-press data. This resulted in the inclusion of data from 15 additional studies.
In total, our dataset included data from S = 79 sources, K = 94 independent samples, and N =
23,461 individuals. All articles can be found in Online Supplement 1A.
Study Coding
Four researchers were responsible for coding all data (see the Online Supplement 2 for
complete coding). Content coding decisions for EPM characteristics (summarized below and
described in greater detail in Online Supplement 1B) involved a high degree of rater judgment.
We therefore calculated agreement indices for coding decisions regarding EPM characteristics.
Consistent with previous meta-analytic efforts (e.g., Van Iddekinge et al., 2019) intercoder
agreement was used as an index of reliability. There was relatively high agreement for coding of
EPM characteristics (intercoder agreement = 82.9% across all coded relationships, and ranged
from 70% agreement for relationships coded as for Surveillance purpose to 93% agreement for
relationships coded as Transparency). All coding decisions, including EPM characteristics, were
compared across coders, with disagreements reconciled through group discussion.
Coding of EPM Characteristics
For studies included in the Presence analysis (studies that included monitored and
unmonitored individuals), we coded for EPM characteristics in each study as between-study
level variables. For studies included in the Degree analysis (studies for which the presence of
EPM was a constant), only EPM characteristics that were directly examined as within-study
variables were coded. EPM purpose was derived from the monitoring rationale provided to
A META-ANALYSIS OF THE EFFECTS OF EPM
24
individuals in a study. When individuals in a study were told that they were monitored for
comparisons against others or some performance standard, the purpose was coded as
Performance; when individuals were told that they were monitored for record keeping purposes
or for their safety, the study was coded as Admin/Safety; when individuals were told that they
were monitored to provide them with developmental feedback or to help them in learning a new
skill, the purpose was coded as Development; and when individuals were told that they would be
monitored with no clear justification provided, the purpose was coded as Surveillance. A small
number of BWC studies with police populations that fit study inclusion criteria were coded as
Admin/Safety. Given the particularly complex nature of police BWC we present results for
Admin/Safety monitoring with and without BWC studies.
Studies with within-study comparisons of monitoring that varied in breadth (e.g., greater
vs lesser variety of monitoring), specificity (e.g., individual monitoring as compared to group
monitoring), target (e.g., task targeted monitoring compared to person targeted monitoring) or
constraints/target control over monitoring (e.g., more or less control over when and where
monitoring takes place and how collected information is used) were coded as invasiveness
broadly and coded for their respective typology sub-elements more specifically. For the Presence
analysis, in which invasiveness was coded at the between-study level, we coded for specificity
and target, but did not code for the breadth or constraints/target control of the monitoring due to
insufficient information for making judgments about what constituted high compared to low
levels of these characteristics at the between-study level.
Synchronicity of collection was coded as either synchronous (performance information is
collected, analyzed, and evaluated in real-time) or asynchronous (performance information is
archived to be viewed or analyzed later). Finally, data extracted from articles that represented the
A META-ANALYSIS OF THE EFFECTS OF EPM
25
relationship between EPM with more or less transparency (e.g., individuals who were given
more or less information about when, how, or why monitoring took place) and individual
outcomes were coded as transparency. There was insufficient information available to make
judgements about what constituted high or low levels of transparency at a between-study level,
and therefore we did not code this variable at a between-study level.
Coding of Work Outcomes
Consistent with previous meta-analytic efforts (e.g., Doerwald et al., 2021; Rudolph et
al., 2017), we used a synthetic construct grouping approach to code for work-related outcome
variables that overlap theoretically and empirically (see Table 1 for variable groupings). We used
the online metaBUS platform (Bosco et al., 2015) to establish evidence for the relationships
between several outcomes in our study. metaBUS is a search engine created to rapidly
summarize research findings. Several outcomes were strongly interrelated according to the
metaBUS database, justifying their combination into a synthetic construct (e.g., job satisfaction
and task satisfaction, r = 0.53; burnout and fatigue, r = 0.55; affective commitment and
organizational commitment, r = 0.51). We drew from theory to combine outcomes with
insufficient data in metaBUS to estimate mean correlations (e.g., privacy concerns and privacy
invasion).
In addition to our synthetic groupings, we content coded task performance criteria. Task
performance was narrowly coded as speed/quantity (e.g., typing speed, response time, data entry
attempts) or as accuracy (e.g., percentage of data entries correct, response accuracy) and left
uncoded when performance required both speed/quantity and accuracy (interrater agreement =
93.6%). We include these narrower performance criteria as subgroups of task performance. We
meta-analyzed relationships between EPM characteristics and synthetic outcomes that appeared
A META-ANALYSIS OF THE EFFECTS OF EPM
26
in at least three (K ≥ 3) independent samples. We also report the relationship between EPM
characteristics and more narrow outcomes within synthetic groupings when K ≥ 3 (e.g., OCBs,
CWBs, burnout). Work outcomes that neither rationally fit within a synthetic construct (e.g.,
number of arrests by police officers; performance variability) nor appeared in K ≥ 3 relationships
were not included in analyses.
-- Insert Table 1 about here --
Coding of Study Method
In addition to understanding the ways that monitoring characteristics influence relevant
work outcomes, it is also important to understand the degree to which research methods used to
study EPM may influence these observed relationships. We therefore coded for study
methodology as a potential moderator. Empirical EPM research tends to be conducted in one of
three ways: 1) experimental studies in which a group of individuals are monitored, or made to
think they are monitored, while they work or perform some task (e.g., Becker & Marique, 2014),
which we coded as “monitoring experiments” ; 2) experimental studies in which individuals are
presented with vignettes or scenarios depicting EPM (e.g., McNall & Roch, 2007), which we
coded as “vignette studies”; and 3) non-experimental survey research in which survey questions
assess experiences with EPM and relevant work outcomes (e.g., Sprigg & Jackson, 2006), which
we coded as “non-experiments”.
2
Meta-Analytic Procedures
All analyses were conducted in R using the psychmeta package (Dahlke & Wiernik,
2019). We corrected observed correlations for sampling and measurement error and estimated
2
We also coded studies as occurring in field or laboratory setting. The large majority of field studies were non-
experiments, whereas lab studies were highly variable (e.g., high fidelity simulations, vignette studies) and thus, we
caution against interpretation of these effects as representing the fidelity of the setting. Results for analyses with
field vs lab setting can be found in Online Supplement 1D.
A META-ANALYSIS OF THE EFFECTS OF EPM
27
meta-analytic effect size estimates using Hunter and Schmidt’s (2004) random effects
procedures. First, we corrected for sampling error by calculating sample size-weighted
correlations. Second, where possible (i.e., for multi-item scales), we corrected for lack of perfect
reliability. Artifact distributions were used when reliabilities for multi-item scales were not
reported (Hunter & Schmidt, 2004). Taking the most conservative approach, we did not correct
for lack of perfect reliability in single-item scales or objective outcomes (e.g., number of words
typed in a data entry task). We also chose not to correct for measurement error in experimental
manipulations, given that variance in the degree to which participants in experimental paradigms
are perfectly certain about if, when, and how they are being monitored, even when explicitly
told, reflects uncertainties about EPM that exist in workplaces as well (Alder, 2001). Next, linear
composites were calculated (Hunter & Schmidt, 2004) when an independent sample reported
multiple correlations for the same relationship between an EPM variable and a synthetic
outcome. For the EPM Presence analysis, EPM characteristics were treated as moderators of the
relationship between EPM presence and work outcomes. For both the Presence and Degree
analysis, study methodology (i.e., monitoring experiment, vignette study, non-experiment) was
also examined as a moderator. We used hierarchical subgroup moderator analysis due to the
relatively small subgroup K (Schmidt, 2017). We additionally conducted PET-PEESE analysis
(Stanley & Doucouliagos, 2014) and cumulative meta-analysis to address the influence of
publication status on our conclusions.
In addition to the sample size-weighted correlation () and the sample size-weighted and
reliability-corrected correlation (), we report the 95% confidence interval and the 80%
credibility interval for . When the confidence interval for does not include zero, it is
considered statistically significant. Confidence intervals can be directly compared across
A META-ANALYSIS OF THE EFFECTS OF EPM
28
different levels of the same moderator, with non-overlapping 95% confidence intervals
suggesting that moderator subgroups are statistically different from one another (p < 0.05). When
the credibility interval is wide moderators are likely present whereas when the credibility interval
is narrow any possible moderators can only have small effects (Geyskens et al., 2009).
-- Insert Table 2 about here --
Results
Meta-analytic results for the relationships between EPM characteristics and work
outcomes are summarized in Table 2-5 for the EPM Presence analysis and Table 6 for the EPM
Degree analysis. Effect sizes found in the EPM Presence analysis (K = 59) represent the absolute
effect of a monitoring characteristics whereas those found in the EPM Degree analysis (K = 44)
represent the effect of greater or lesser amounts of a characteristic. All relationships that met our
K ≥ 3 criterion were included in the results; however, in some cases, there was not sufficient K to
make outcome comparisons across all EPM characteristics. Readers should interpret results from
analyses with small Ks with caution. However, as Valentine et al. (2010) note, even when K = 2,
meta-analysis is superior to other means of synthesis.
Overall Effects of EPM
Prior to testing our hypotheses, we examined the overall effects of EPM on outcomes
without including the coded monitoring characteristics in analyses. As expected, results from
these analyses were not predictive of work outcomes and strongly suggested the presence of
moderators for nearly all relationships. As shown in Table 2 no statistically significant effects
were observed between the mere presence of monitoring and performance, or any subset of
performance behaviors (e.g., task performance, contextual performance)
3
. Similarly, the
3
A reviewer raised the possibility that null relationships found between EPM and performance could be attributable
to studies measuring non-relevant or very distal performance criteria. For instance, in a study in which employee
A META-ANALYSIS OF THE EFFECTS OF EPM
29
confidence intervals for nearly all relationships between EPM and attitudes were wide and all
relationships except EPM with broad attitudes ( = -.11 [95%CI: -.20; -.03]) and privacy
invasion ( = .28 [95%CI: .13; .43]) contained 0. Further, the credibility intervals for all
relationships between EPM and performance and attitudinal outcomes were wide, indicating the
likely presence of moderators.
The relationship between EPM and stress/strain was positive and significant ( = .16
[95%CI: .13; .20]), with a relatively narrow credibility interval. No studies of Development EPM
measured stress or strain outcomes. There were sufficient samples of other EPM purposes
included in analyses however, as well as variation in EPM target, and synchronicity of
collection. As can be seen in Tables 3-5, all levels of purpose, invasiveness, and synchronicity
were positively and significantly related to stress/strain, and no level of the moderators
significantly differed from each other. These results suggest that, independent of monitoring
characteristics, the presence of EPM may tend to increase worker stress. Together, these initial
analyses demonstrate the need for further differentiation of EPM studies by EPM characteristics.
-- Insert Table 3 about here --
EPM Purpose
We hypothesized that the articulated purpose of EPM moderates the effect of EPM on
performance (H1a), attitudes (H1b) and stress/strain (H1c). We tested these hypotheses with the
EPM Presence analysis, examining the coded purpose of monitoring as subgroup moderator of
the effects of presence/absence of EPM. As shown in Table 3, we found little support for these
emails were monitored to catch phishing attempts, one might expect a weak relationship between EPM and overall
job performance, and this weak effect could mask a relationship between EPM and more proximal or relevant
performance criteria. We therefore reviewed each study included in our meta-analysis that measured performance
for non-relevant or distal performance criteria in relation to monitoring. No such studies were identified and
therefore, we did not conclude that the null relationship was due to the use of non-relevant performance criteria.
A META-ANALYSIS OF THE EFFECTS OF EPM
30
hypotheses. For H1a, estimates for the moderating effects of EPM purpose on performance were
all non-significant, with effect sizes near zero. There was an insufficient sample of studies (K= 2)
that examined the effects of Development EPM on performance. Similarly, little support was
found for H1b. In general, the effects of monitoring on attitudes did not differ by purpose.
Additionally, credibility intervals for nearly all relationships between the four purposes and
attitudinal outcomes were wide, indicating the likely presence of moderators. As previously
described, results did not show a moderating effect of EPM purpose on stress/strain. Thus, H1c
was not supported.
--Insert Table 4 & 5 about here --
Invasiveness
We predicted that invasiveness moderates the effects of EPM on performance such that
more invasive EPM is associated with more positive performance (H2a). We had sufficient data
to test aspects of EPM Invasiveness in both the EPM Presence Analysis (Table 4) and the Degree
Analysis (Table 6) Results, for the most part, did not suggest that EPM with varying levels of
invasiveness have significant effects on performance, with one notable exception. As shown in
Table 6, results indicated that EPM of greater breadth (i.e., monitoring individuals in more ways)
significantly and positively predicted counterproductive work behaviors (CWBs; e.g., antisocial
behavior, withholding effort) ( = .22 [95%CI: .005; .440]). Beyond this observed relationship,
invasiveness broadly, nor the sub-elements of invasiveness (e.g., target, breadth,
constraints/target control), were not found to predict performance behaviors. There was an
insufficient sample of primary studies (K = 2) that directly compared the effects of EPM with
more or less specificity to analyze the effects of specificity on its own, but specificity
correlations were included in the broader analysis of EPM Invasiveness.
A META-ANALYSIS OF THE EFFECTS OF EPM
31
We predicted that invasiveness moderates the effects of EPM on attitudes such that more
invasive forms of EPM are associated with more negative work attitudes (H2b). In our Degree
Analysis, invasiveness (i.e., EPM with greater breadth, more personal target, more specificity, or
less constraints/target control) negatively and significantly related to broad attitudinal outcomes
( = -.17 [95%CI: -.22; -.11]), fairness and justice perceptions ( = -.14 [95%CI: -.25; -.02]),
commitment ( = -.18 [95%CI: -.36; -.01]), and autonomy ( = -.18 [95%CI: -.26; -.09]) and
positively and significantly related to privacy invasion ( = .20 [95%CI: .10; .29]). In terms of
the specific sub-elements of invasiveness, breadth was significantly and negatively related to
perceptions attitudes broadly ( = -.18 [95%CI: -.25; -.11]) and autonomy ( = -.18 [95%CI: -
.30; -.05]) and positively related to privacy invasion ( = .23 [95%CI: .09; .37]). Providing
greater constraints/target control over EPM positively and significantly related to broad
attitudinal outcomes ( = .17 [95%CI: .08 .25]), and more narrowly, positively related to
perceptions of fairness and justice ( = .27 [95%CI: .10: .45]) and autonomy ( = .22 [95%CI:
.12: .31]) and negatively related to privacy invasion ( = -.20 [95%CI: -.32; -.08]). In the
Presence analysis, task-targeted EPM (less invasive) and person-targeted EPM (more invasive)
did not differ in their effects on broad or specific attitudes. However, as seen in Table 4, a pattern
emerged such that estimates for person-targeted EPM tended to be larger than those for task-
targeted EPM. For instance, the relationship between person-targeted EPM and privacy invasion
was significant and negative for person-targeted EPM but not for task-targeted EPM. Overall, the
above findings support H2b.
We predicted that invasiveness moderates the effects of EPM on stress/strain such that
more invasive forms of EPM are associated with greater levels of stress/strain (H2c). Results
from the Degree Analysis indicated that more invasive EPM was associated with greater levels
A META-ANALYSIS OF THE EFFECTS OF EPM
32
of stress/strain ( = .20 [95%CI: .10: .29]). At the sub-element level, breadth was positively
related to stress/strain outcomes ( = .22 [95%CI: .12: .31]), including burnout ( = .18 [95%CI:
.06: .30]). Too few studies examined the relationship between constraints/target control and
stress/strain (K = 2) to include in analyses. Results from the Presence Analysis, however, did not
show that task-targeted and person-targeted monitoring differed in their effect on stress/strain.
Thus, partial support was found for H2c.
Synchronicity of Collection
We asked whether synchronicity moderates the effect of EPM on performance (RQ1a),
attitudes (RQ1b), and stress/strain (RQ1c). We were able to examine this research question with
data included in the Presence Analysis. Results provided mixed evidence for differences between
synchronous and asynchronous EPM (see Table 5). Synchronicity was found to moderate the
effect of EPM on broad attitudes, such that the effect was negative when EPM was synchronous
( = -.29 [95%CI: -.37; -.21]) but not when EPM was asynchronous ( = .00 [95%CI: -.19; .19]).
We did not find evidence, however, for synchronicity as a moderator of the effects of EPM on
more narrowly defined attitudes nor did we find evidence for synchronicity as moderator the
effects of EPM on performance or on stress/strain.
Transparency
We predicted that transparency moderates the relationship between EPM and
performance, attitudes, and stress/strain such that greater transparency is associated with more
positive performance (H3a) and attitudes (H3b) and lower levels of stress/strain (H3c). We were
able to test two of these hypotheses with data in the Degree analysis. Results did not show that
transparency had any effect on performance overall, or task performance or contextual
performance more narrowly. Thus, support was not found for H3a. As shown in Table 6,
A META-ANALYSIS OF THE EFFECTS OF EPM
33
transparency was found to positively relate to attitudes overall ( = .25 [95%CI: .11; .39]). More
narrowly, transparency was positively related to fairness and justice ( = .21 [95%CI: .04; .38]),
commitment ( = .35 [95%CI: .10; .61]), perceived support ( = .37 [95%CI: .13; .62]), and
satisfaction ( = .26 [95%CI: .08; .44]). Overall, these findings support H3b. There was an
insufficient sample (K=2) to test H3c regarding the moderating effects of transparency on
stress/strain.
--Insert Table 6 about here--
Study Method as a Moderator
Including study method (monitoring experiment, vignette experiment, non-experiment)
did not reveal significant method effects in most cases. One exception was in the relationship
between EPM and attitudes. Participants in vignette-based experiments reported significantly
more negative attitudes ( = -.32 [95%CI: -.46; -.19]) than those in non-experiments ( = .00
[95%CI: -.15; .16]).
Robustness and Post Hoc Analysis
Sensitivity analyses did not provide evidence that publication bias affected estimates in
our study. More specifically, the slope of the standard error for the relationship between EPM
and tested outcomes in PET-PEESE models were all nonsignificant, and cumulative meta-
analysis in which published and unpublished studies sequentially entered into analysis showed
no evidence of “drift” (McDaniel, 2009). See Online Supplement 1C for more detailed findings
from sensitivity analyses.
One possible explanation for the null findings regarding the moderating effect of EPM
purpose could be heterogeneity within categories of purpose. In particular, the category of
Performance EPM included a variety of specific purposes. We identified three sub-categories
A META-ANALYSIS OF THE EFFECTS OF EPM
34
within Performance EPM: 1) studies in which the purpose of EPM was to deter CWBs such as
theft or off-task behaviors (recoded as Deterrence EPM); 2) studies in which the purpose of
EPM was to help inform performance related decisions such as performance evaluations
(recoded as Evaluative EPM); and 3) studies in which the purpose was to provide specific
rewards or punishments (recoded as Incentive EPM). All studies were coded by 2 researchers
(Cohen’s kappa = .76). Meta-analytic estimates were re-calculated including the coded
subgroups as moderators of relationships with sufficient K for such analysis. Results did not
show any significant differences between the newly coded subgroups (see Online Supplement
1D). Therefore, we did not find evidence that heterogeneity within the Performance EPM
operationalization was responsible for null findings.
DISCUSSION
Though the COVID-19 pandemic has presented organizations with new concerns about
EPM, the practice of electronically monitoring workers has existed for decades. Recent media
reports suggest that sales of EPM services have risen dramatically since the pandemic began
(Golden & Chemi, 2020), perhaps because organizations are implementing EPM to track
individuals as they work remotely (Bureau of Labor Statistics, 2020; Satariano, 2020). Further,
electronic forms of safety monitoring (e.g., location tracking to ensure physical distancing, body
temperature tracking) may be required as individuals return to workplaces. These practices are
indeed new for many organizations, but the principles of applied psychology and available
scientific evidence are still essential to their implementation.
The present study extends previous reviews of the EPM literature (e.g., Ravid et al.,
2020; Stanton, 2000) by applying meta-analytic techniques to reveal patterns of relationships
among EPM and a variety of work-related outcomes. Specifically, we present empirical evidence
A META-ANALYSIS OF THE EFFECTS OF EPM
35
to guide the monitoring decisions of organizations in ways that consider effects on worker
attitudes, performance, and stress. While some of our findings are consistent with observations
from Ravid and colleague’s (2020) qualitative review (e.g., the negative effect of invasiveness on
attitudes), other findings (e.g., the null effects of EPM on performance) deviate from their
review. Thus, the current study presents an opportunity to advance psychological theory and
organizational practice of EPM in an evolving world of work.
Summary and Interpretation of Findings
There were several noteworthy findings in this study. First, the presence of EPM
increases stress/strain, regardless of monitoring characteristics. This conclusion is strengthened
by the variety of stress indicators (e.g., fatigue, performance pressure, anxiety, physiological
indicators) included in the analysis and the consistent and moderate effect sizes observed. Ravid
and colleagues (2020) briefly discuss the effects of EPM on worker stress in their review, but
these effects are worthy of further consideration. Our results are in line with Zajonc’s (1965)
assertion that the mere presence of others (in this case, regardless of monitoring characteristic)
increases arousal levels and suggest that EPM should be treated as a job demand (Demerouti et
al., 2001) with associated physiological and psychological costs. There were no studies
examining Development EPM in our analyses of stress/strain, however; it remains possible that
workers feel significantly less stressed when they perceive EPM to be for their own growth and
development, and future research should examine this possibility.
Second, relative levels of EPM characteristics do influence work outcomes—particularly
regarding the effects of invasiveness and transparency on attitudes. Invasiveness was found to
moderate the effects of EPM on attitudes such that more invasive EPM was generally associated
with more negative attitudes. No observed relationship in our meta-analysis was stronger than
A META-ANALYSIS OF THE EFFECTS OF EPM
36
the relationship between person-targeted EPM and perceptions of privacy invasion, indicating
that tracking of more personal kinds of information (e.g., location, social exchanges) is
particularly likely to violate perceived privacy expectations. The privacy calculus model argues
that workers react negatively when the perceived costs of giving up their privacy outweigh the
benefits (Acquisti, 2009). As EPM becomes more invasive, individuals forgo more of their
privacy, thus raising the costs. The observed positive relationship between EPM Invasiveness
and CWBs (e.g., withholding effort, computer abuse) in our study hints at how individuals may
respond to these rising costs. The positive relationship between invasiveness and CWBs was
particularly noteworthy given that the explicit purpose of monitoring is often to deter CWBs.
Thus, highly invasive EPM may be counterproductive in many circumstances.
Findings regarding the moderating effects of EPM Transparency provide further
information about how workplaces may mitigate negative attitudinal effects of EPM. Per the
privacy-calculus model, individuals may react more favorably to EPM when they understand
how the system will benefit them or their organization (Acquisti, 2009). As such, organizations
using invasive forms of EPM should clearly communicate the reasons that they are collecting
personal information. Our study demonstrates the benefits of greater transparency regarding
monitoring decisions—workers report more satisfaction, commitment, and perceived support,
and more positive fairness perceptions, when their organizations are open about EPM.
Transparency becomes even more important as monitoring technologies become increasingly
discreet and unobtrusive, as there is more opportunity to conceal what is being tracked.
An unexpected finding was how little the effects of monitoring appeared to differ based
on the explicit purpose communicated to workers. Due to the state of EPM scholarship, we
offered very broad hypotheses regarding the moderating effects of purpose, and yet, were still
A META-ANALYSIS OF THE EFFECTS OF EPM
37
unable to find evidence to support these hypotheses. There are several possible explanations for
these counterintuitive findings. It is possible that individuals often come to their own conclusions
about the purpose of monitoring even when a purpose is communicated to them. They may also
assume that monitoring occurs for multiple purposes at once. The few studies that included
manipulation checks lend some credence to this idea. Karim (2015) reported that participants’
assignment into a development EPM condition (versus administrative EPM condition) was
associated with perceptions that the EPM was used for administrative, not development,
purposes. Wells and colleagues (2007) similarly found that workers in a single organization had
varying perceptions regarding whether their shared EPM system was used for developmental and
deterrent purposes. Our null findings suggest the possibility that the way EPM is psychologically
experienced is more important than what an organization communicates to their employees.
We also found little evidence for a moderating effect of study method (i.e., monitoring
experiment, vignette study, non-experiment), suggesting that a variety of methods may be
appropriate for studying EPM. These null findings were surprising given evidence that
individuals’ self-reported hypothetical behaviors are often misaligned with their actual behaviors
(Baumeister et al., 2007). Studies on affective forecasting have shown that individuals tend to
overreact to imagined negative events more than they do to imagined positive events (Wilson &
Gilbert, 2003). Thus, there are theoretical reasons to think that experiencing EPM with real
stakes will differ from imagining how one would experience EPM in a hypothetical scenario. We
encourage future research to extend our findings regarding study method.
Finally, we found little evidence for the effects of monitoring and monitoring
characteristics on performance behaviors. These null findings deviate from the conventional
wisdom that EPM increases motivation to perform. Small or non-significant effects were
A META-ANALYSIS OF THE EFFECTS OF EPM
38
observed across a variety of performance criteria (e.g., speed/quantity, accuracy, performance
complaints, and contextual performance). Comparing these null findings to the positive
performance effects observed in meta-analyses of other work practices such as mentoring (Eby et
al., 2008), coaching (Theeboom et al., 2014), and multisource feedback (Smither et al., 2005),
we can conclude that EPM appears to be a relatively ineffective intervention if one’s goal is
improving worker performance.
Implications for the Typology of EPM Characteristics
The current study was the first to test the usefulness of Ravid and colleagues’ (2020)
typology of EPM characteristics as a framework for predicting work outcomes. Results show that
some parts of the typology, such as transparency and invasiveness, were diagnostically useful for
work outcomes (e.g., privacy invasion, justice perceptions) and other aspects, such as purpose
and synchronicity, were less useful. Our results suggest that focusing on a single EPM
characteristic (i.e., purpose, synchronicity) may not fully clarify the relationships between EPM
and work outcomes, perhaps because of high levels of heterogeneity. We encourage future
research that examines interactive effects of EPM characteristics. For instance, any positive
performance effects of invasiveness may only be observed when considered in combination with
the synchronicity of feedback. Highly invasive EPM may be able to capture performance
behaviors in great detail and specificity, but these benefits may be lost if feedback is not timely
(DeNisi & Kluger, 2000; Lechermeier & Fassnacht, 2018).
We also encourage researchers to continue to evolve the typology of EPM characteristics
as new technologies enter workplaces and become the focus of research. For instance, as a body
of research develops regarding the use of EPM for health and safety purposes, it may become
necessary to distinguish health and safety monitoring purposes from administrative and liability
A META-ANALYSIS OF THE EFFECTS OF EPM
39
purposes. Likewise, one generally unexplored purpose for monitoring is that of acting as a coach
or recommender to support workers. We identified one study of such monitoring: Carayon
(1994) investigated office workers’ reactions to EPM used for phone call routing. The EPM
system in that study tracked workers’ calls to inform them of their call acceptance volume and
transfer calls to workers with lower volume. We coded this study as Admin/Safety EPM, but we
believe, a new category, Support, may more accurately represent the purpose and will be a
significant category moving forward. As more EPM tools are designed and implemented with
job design and organizational development in mind (Landers & Marin, 2021), Support EPM will
present research opportunities to understand how workers structure and make sense of their jobs.
--Insert Table 7 about here --
Directions for Future Research
Results highlight a number of avenues for future EPM research. We discuss some of
these opportunities in the following section and present further opportunities in Table 7.
Unstudied and Understudied Areas from the Typology
There were several aspects of the EPM typology that we were unable to test due to
absence of primary studies; other characteristics were included in the meta-analysis but can be
regarded as understudied (i.e. K 5). For example, we were unable to include synchronicity of
feedback in either of our analyses. A key distinction between traditional and electronic
monitoring is in the capability of EPM to synchronously collect and analyze performance data
and deliver almost instantaneous performance feedback. Research is needed to understand how
real-time synchronous work feedback may influence performance, attitudes, and stress.
By far the most prevalent EPM purpose we identified was Performance, with fewer
studies examining EPM for other purposes. We were able to identify few studies that examined
A META-ANALYSIS OF THE EFFECTS OF EPM
40
EPM for health and safety purposes (i.e., Dumlao et al., 2019; Raveendhran & Fast, 2021),
despite increasing production and marketing of monitoring technologies designed to keep
employees safe and reduce worksite accidents (Mamun & Yuce, 2019). The COVID-19
pandemic has raised further questions about how employees experience and respond to health
and safety monitoring in times of emergency. The affordability of wearable devices capable of
collecting health indicators makes such research practically relevant and feasible. There is a clear
need for research that explores how individuals respond to EPM for health and safety and the
contextual and environmental factors that may change these responses.
We identified very few studies examining the effects of EPM for training and
development. Given the rapid increase in virtual and augmented reality technologies that
organizations are implementing to track, train, and develop employees (i.e., Development EPM;
Rogers, 2020), it is important that future research examine how Development EPM influences
workers. The question of which job skills are most appropriate for delivering developmental
feedback via EPM, for example, is unexplored. There is also much to learn about the
circumstances under which individuals trust and incorporate algorithmic feedback captured via
EPM as compared to human feedback, and what EPM design elements may alter these
perceptions? Answers to these questions have implications for organizations implementing
training technologies as well as for designers of those technologies. Finally, there were
insufficient primary studies that examined EPM technologies that targeted individuals’ thoughts,
feelings, and physiology (e.g., social media monitoring, physiological monitoring), which might
be considered the most invasive forms of monitoring (Kaupins & Coco, 2017) and can
increasingly be found in workplaces (Morris et al., 2017).
Purpose Perceptions
A META-ANALYSIS OF THE EFFECTS OF EPM
41
We found little evidence that the communicated purpose of monitoring moderates the
effects of EPM on work outcomes. Post-hoc analyses did not suggest that these null findings
were due to heterogeneous sub-categories of purpose (e.g., Deterrence, Incentives). These null
findings suggest the possibility that purpose may be better explored as a set of multi-dimensional
perceptions. We encourage the development of multi-dimensional scales of EPM purpose
perceptions to better capture interactive effects of EPM purpose. Such studies may also allow for
the discovery of profiles of monitoring purpose (e.g., EPM that is strongly perceived as for
Performance and Development as distinct from EPM that is strongly perceived as for both
Performance and Safety). Perceptions of EPM purpose no doubt evolve and shift over time,
across contexts, with cultural shifts, and with experience. Many of the studies included in our
meta-analysis conveyed the purpose of monitoring shortly before collecting outcome measures.
It is possible that effects of purpose may only be observed once individuals are convinced of the
purpose, requiring time, repeated exposure, and trustworthy organizational communication.
Boundary Conditions and Mediators
Interpersonal and contextual moderators may help explain the significant heterogeneity
observed in our results. For instance, although some research has examined trust in management
as an outcome of EPM (e.g., Butler, 2012; McNall & Roch, 2009) it is also likely to be an
important determinant of individual responses to monitoring. In the current study, EPM
characteristics such as Purpose were coded according to the nature of explicit communications to
workers, but these communications will only have predictable effects to the degree that
employees believe that they are truthful. Individuals in organizations, as well as those
participating in research studies, who have low levels of trust in the organization (Rousseau et
al., 1998) may come to their own conclusions about how and why monitoring is occurring.
A META-ANALYSIS OF THE EFFECTS OF EPM
42
Similarly, the positive effects of transparency may be attenuated when trust in an organization is
very high, as employees may assume their organization is acting in good faith even with little
information provided (Bhattacharya et al., 1998). We see interpersonal and organizational trust
as a potentially important boundary condition for future researchers to explore.
There is also need for a better understanding of how individuals perceive and receive
work monitoring in their home or personal space. Very few studies included in our meta-analysis
captured the work setting in which monitoring took place (e.g., work office, home office, during
commutes, during work breaks). Wang and colleagues (2021) recently observed employee
monitoring as a prominent theme emerging from discussions with individuals who transitioned to
remote work during the pandemic, reporting that the number of ways that individuals were
monitored in their home predicted greater levels of work-to-home interference. Monitoring that
is seen as appropriate in a centralized work location may be seen as overly invasive elsewhere.
Finally, a number of mediating mechanisms are likely to exist, through which EPM
influences work outcomes. Exploring privacy invasion perceptions as a mediator between EPM
characteristics and work outcomes of interest may be a particularly promising route. Such
research would also answer recent calls (e.g., Bhave et al., 2020) for studies to better understand
how individuals evaluate and respond to workplace practices that have the potential to violate
privacy expectations.
Unexamined Outcomes of EPM
Several work outcomes were measured in too few studies to be included in our analysis
but should be explored in future work. The majority of studies in our analysis measuring
performance focused on tasks such as data entry and customer support that have easily
measurable performance criteria such as task speed or quantity of output. As EPM becomes more
A META-ANALYSIS OF THE EFFECTS OF EPM
43
varied in design and use, there is a need for research examining other dimensions of
performance. We were unable to identify research examining the effects of monitoring on safety
behaviors, although these are probably of primary interest for those using Admin/Safety EPM.
One study included in our meta-analysis (Schlund, 2020) measured the effect of Performance
EPM on objective creativity criteria (i.e., the number of ideas generated in a creativity task) but
further work is needed to better understand the effect monitoring on subjective creativity (e.g.,
the quality of the ideas), particularly as creative team processes increasingly move into
monitored virtual spaces (Thompson, 2020).
There is a need for research exploring the effects of EPM on recovery experiences shown
to relate to important work and health outcomes (Sonnentag & Fritz, 2007). We found that the
presence of monitoring was associated with greater levels of stress/strain. This finding raises
questions about how the presence of these technologies influence individuals during breaks.
Might the mere presence of monitoring technologies negatively affect one’s ability to experience
relaxation during a rest break? How might monitoring influence one’s ability to psychologically
detach - the subjective feeling of leaving work behind – during breaktimes? And how might
monitoring technologies (e.g., physiological monitoring) be implemented to assist in facilitating
these important experiences?
We also see a need for research examining the effects of EPM on team and group level
processes. As work shifts to more virtual formats, teamwork increasingly takes place virtually
(Meluso et al., 2020). This shift may result in fewer available channels for non-monitored
teamwork. Evaluation expectations have been shown to alter group processes (e.g., Amabile et
al., 1990). Research is needed to better understand the effects of EPM on team-level processes
A META-ANALYSIS OF THE EFFECTS OF EPM
44
such as communication and member integration, and whether the characteristics of EPM
moderate these effects.
Practical Implications
The current study provides important practical guidance for organizations considering the
implementation of EPM. First, we caution organizations against investing in EPM with the
expectation of guaranteed worker performance improvement. Many EPM systems represent a
significant financial investment (Zielinski, 2020) with an expectation that these costs will
translate to quick improvements in performance. Our study found no such effects.
Second, we recommend that organizations that choose to electronically monitor workers
do so in ways that are minimally invasive. We found that more invasive monitoring was
associated with a number of negative attitudinal outcomes, as well as increased reports of CWBs
and stress, without any evidence for performance improvement. We therefore advise leaders to
make decisions about what and how to monitor as conservatively as possible, aiming to monitor
narrowly (e.g., capturing only task relevant information when possible) and limiting the tracking
of more personal kinds of information to circumstances where such invasiveness is merited (e.g.,
to ensure worker safety). Setting clear parameters about how collected data can and cannot be
used and finding ways to provide employees with greater levels of control over monitoring (e.g.,
allowing employees to turn monitoring off during work breaks) can further mitigate negative
attitudinal effects (Alge et al., 2006; Slemp et al., 2018).
Third, individuals find monitoring to be stressful. As such, EPM should be considered a
work demand that requires effort expenditure and corresponding opportunities for recovery
(Meijman & Mulder, 1998). Just as individuals benefit from formal and informal breaks (e.g.,
lunchbreaks, coffee breaks) from other work demands (Sonnentag et al., 2017), individuals are
A META-ANALYSIS OF THE EFFECTS OF EPM
45
likely to benefit from breaks from monitoring. We recommend that organizations using EPM
provide workers with periods of time throughout a work shift in which monitoring is greatly
reduced or removed all together as times for workers to rest and recover. This advice is
particularly important as it becomes easier for organizations to monitor workers continuously
and in a variety of ways throughout work shifts, including during rest breaks.
Fourth, maximizing transparency when using EPM is essential (e.g., informing workers
how and when monitoring will occur, what data will be collected and who will have access to
them) to minimize negative work attitudes in monitored individuals (Lowry et al., 2015; McNall
& Roch, 2007). As noted above, beliefs about monitoring purpose often differ from what is
officially communicated. We recommend that organizations have consistent conversations with
employees about the purpose of monitoring, clearly connect this purpose to organizational and
individual goals and values, and limit monitoring to ways consistent with the communicated
purpose to convince workers that alternative motives for monitoring do not exist (Mayer &
Davis, 1999; Schnackenberg & Tomlinson, 2016).
Implied in each of these recommendations is a need for organizations to first identify
behaviors that describe performance and then find appropriate ways to measure those behaviors,
rather than the other way around. Although EPM makes it easy to measure a wide range of
behaviors, organizations can expect favorable work outcomes only when the characteristics of
their EPM systems adhere to principles of applied psychology. EPM is just one of many work
practices influencing workers at any given time; as such, even small effects may be practically
meaningful. Effect sizes observed in the current analysis range from small to modest in strength
and are similar in strength to those reported in meta-analyses of other important work
interventions such as high-performance work practices (Combs et al., 2006) and
A META-ANALYSIS OF THE EFFECTS OF EPM
46
telework (Gajendran & Harrison, 2007). We believe that results from the current study provide
practical guidance for organizations considering using EPM.
Limitations
We note several limitations to this analysis. First, meta-analyses are always constrained
by the population of studies available for a given research question. Several aspects of the EPM
typology could not be tested due to absence of primary studies (K < 3) and caution should be
exercised when interpreting results that are based on a relatively small number of primary studies
(e.g., K < 5). However, as argued by Schmidt, “even with a small number of studies and small
ns, meta-analysis is the optimal method for integrating findings across studies” (Schmidt et al.,
1985, p. 749). We hope the current study serves to guide future research towards aspects of EPM
that remain un- or understudied.
Second, although our synthetic construct approach was necessary to accumulate a
sufficient sample of work outcomes to conduct a comprehensive meta-analysis, this compound
approach may produce greater between-study variance in outcomes, and thereby larger standard
errors and inferential uncertainty. Thus, the wide confidence intervals and non-significant results
that were observed in this study are likely, in part, due to the variance attributable to
measurement. Similarly, characteristics within the EPM typology are theoretically derived, but
prior to the current study, empirically untested. It is possible that heterogeneity exists within the
proposed characteristics of monitoring that contributed to wide confidence intervals. Our post-
hoc examination of sub-groups within the Performance purpose characteristic did not support
such interpretation, but we encourage future research to continue to examine the proposed
typology to test the degree to which categories describe relatively homogeneous constructs.
A META-ANALYSIS OF THE EFFECTS OF EPM
47
Third, the current study represents a first step in disaggregating the effects of EPM
characteristics but, both in research and practice, certain EPM characteristics are very likely to
covary. For instance, EPM used for Surveillance purposes is likely to be low in transparency
whereas EPM used for Development purposes may tend to be characterized by high levels of
transparency. Performance EPM may tend to be high in Synchronicity of collection to enable
immediate feedback. The current study had limited power to control for these covarying effects
or examine the interactions among EPM characteristics. Fourth, our method of coding EPM
purpose according to the explicit purpose provided to those being monitored was neither able to
capture the perceptions of those being monitored, nor the possibility that monitoring may be
used, or perceived as used, for multiple purposes.
Finally, meta-analyses can broadly be criticized for either being reflective of the quality
of the research that is available in the literature or dependent upon significant findings that have
been published (i.e., the “file drawer” problem). Many studies included in our analysis measured
EPM and outcome variables with a single method. This was more common in studies included in
EPM Degree analysis, perhaps accounting for the stronger effect sizes observed in this analysis.
We expect, however, that as EPM continues serving an important role in modern day work, the
quality and creativity of EPM research that is published will only increase. Considering the latter
issue of the file drawer problem more directly, we took several steps to locate and include
unpublished data sources in our analyses and our publication status sensitivity analysis analyses
did not provide evidence that publication status had a significant effect on our results.
Conclusion
Advances in information technologies and a mass transition to remote work is causing
organizations to monitor employees in myriad ways and with great intensity and detail. A
A META-ANALYSIS OF THE EFFECTS OF EPM
48
growing number of organizations have begun developing and using monitoring software to track
worker behavior (Dreyfuss, 2020). Despite such technological advances, we found no evidence
for EPM as an effective performance intervention. Our meta-analysis suggests that best practices
in human resource management such as honesty, procedural transparency, and providing
individuals with control over their work continue to be important mechanisms for guiding
workers towards individual and organizational goals.
Data Availability Statement
The data that supports the findings of this study are available in the supplementary
material of this article. Online supplements can be found using the following link:
https://osf.io/kc89f/?view_only=78f770d10b704dd3ae91190720ce00b8
A META-ANALYSIS OF THE EFFECTS OF EPM
49
References
Acquisti, A. (2009). Nudging privacy: The behavioral economics of personal information. IEEE
Security and Privacy, 7(6), 82–85. https://doi.org/10.1109/MSP.2009.163
Adams, I., & Mastracci, S. (2018). Police body-worn cameras: Effects on officers’ burnout and
perceived organizational support. Police Quarterly, 22, 5–30.
https://doi.org/10.1177/1098611118783987
Aguinis, H., & Bradley, K. J. (2014). Best practice recommendations for designing and
implementing experimental vignette methodology studies. Organizational Research
Methods, 17(4), 351–371. https://doi.org/10.1177/1094428114547952
Aguinis, H., Pierce, C. A., Bosco, F. A., Dalton, D. R., & Dalton, C. M. (2011). Debunking
myths and urban legends about meta-analysis. Organizational Research Methods, 14(2),
306–331. https://doi.org/10.1177/1094428110375720
Aiello, J. R., & Svec, C. M. (1993). Computer monitoring of work performance: Extending the
social facilitation framework to electronic presence. Journal of Applied Social Psychology,
23(7), 537–548. https://doi.org/10.1111/j.1559-1816.1993.tb01102.x
Ajunwa, I., Crawford, K., & Schultz, J. (2017). Limitless worker surveillance. California Law
Review, 105(3), 735–776. https://doi.org/10.15779/Z38BR8MF94
Alder, G. S. (2001). Employee reactions to electronic performance monitoring: A consequence
of organizational culture. Journal of High Technology Management Research, 12(2), 323–
342. https://doi.org/10.1016/S1047-8310(01)00042-6
Alder, G. S., & Tompkins, P. K. (1997). Electronic performance monitoring: An organizational
justice and concertive control perspective. Management Communication Quarterly, 10(3),
259–288. https://doi.org/10.1177/0893318997010003001
A META-ANALYSIS OF THE EFFECTS OF EPM
50
Alge, B. J., Ballinger, G. A., & Green, S. G. (2004). Remote control: Predictors of electronic
monitoring intensity and secrecy. Personnel Psychology, 57(2), 377–410.
https://doi.org/10.1111/j.1744-6570.2004.tb02495.x
Alge, B. J., Ballinger, G. A., Tangirala, S., & Oakley, J. L. (2006). Information privacy in
organizations: Empowering creative and extrarole performance. Journal of Applied
Psychology, 91(1), 221–232. https://doi.org/10.1037/0021-9010.91.1.221
Alge, B. J., & Hansen, S. D. (2014). Workplace monitoring and surveillance research since
“1984”: A review and agenda. In M. D. Coovert & L. F. Thompson (Eds.), SIOP
Organizational Frontiers Series: The Psychology of Workplace Technology (pp. 209–237).
Routledge/Taylor & Francis Group. https://doi.org/10.4324/9780203735565
Amabile, T. M., Goldfarb, P., & Brackfield, S. C. (1990). Social Influences on Creativity:
Evaluation, Coaction, and Surveillance. Creativity Research Journal, 3(1), 6–21.
https://doi.org/10.1080/10400419009534330
Arnaud, S., & Chandon, J.-L. (2013). Will monitoring systems kill intrinsic motivation? An
empirical study. Revue de Gestion Des Ressources Humaines, 90(4), 35.
https://doi.org/10.3917/grhu.090.0035
imes.com/2017/07/25/technology/microchips-wisconsin-company-employees.html
Bartels, L. K., & Nordstrom, C. R. (2012). Examining big brother’s purpose for using electronic
performance monitoring. Performance Improvement Quarterly, 25(2), 65–77.
https://doi.org/10.1002/piq.20140
Baumeister, R. F., Vohs, K. D., & Funder, D. C. (2007). Psychology as the science of self-
reports and finger movements: Whatever happened to actual behavior? Perspectives on
Psychological Science, 2(4), 396–403. https://doi.org/10.1111/j.1745-6916.2007.00051.x
A META-ANALYSIS OF THE EFFECTS OF EPM
51
Becker, T. E., & Marique, G. (2014). Observer effects without demand characteristics: An
inductive investigation of video monitoring and performance. Journal of Business and
Psychology, 29(4), pp 541–553. https://doi.org/10.1007/s10869-013-9338-1
Benoit, W. L. (1995). Accounts, excuses, and apologies: A theory of image restoration discourse.
SUNY Press.
Bhattacharya, R., Devinney, T. M., & Pillutla, M. M. (1998). A formal model of trust based on
outcomes. Academy of Management Review, 23(3), 459–472.
https://doi.org/10.5465/AMR.1998.926621
Bhave, D. P., Teo, L. H., & Dalal, R. S. (2020). Privacy at work: A review and a research agenda
for a contested terrain. Journal of Management, 46(1), 127–164.
https://doi.org/10.1177/0149206319878254
Bosco, F., Steel, P., Oswald, F., Uggerslev, K., & Field, J. (2015). Cloud-based meta-analysis to
bridge science and practice: Welcome to metaBUS. Personnel Assessment and Decisions,
1(1), 2. https://doi.org/10.25035/pad.2015.002
Brehm, S. S., & Brehm, H. W. (1981). Psychological reactance: a theory of freedom and
control. Academic Press Inc.
Bureau of Labor Statistics. (2020, September 4). Supplemental data measuring the effects of the
coronavirus (COVID-19) pandemic on the labor market. Labor Force Statistics from the
Current Population Survey. https://www.bls.gov/cps/effects-of-the-coronavirus-covid-19-
pandemic.htm
Butler, A. (2012). The effects of organizational justice perceptions associated with the use of
electronic monitoring on employees’ organizational citizenship and withdrawal behaviours:
A social exchange perspective [Unpublished dissertation, University of Windsor].
A META-ANALYSIS OF THE EFFECTS OF EPM
52
https://scholar.uwindsor.ca/etd/478/
Carayon, P. (1994). Effects of electronic performance monitoring on job design and worker
stress: Results of two studies. International Journal of Human-Computer Interaction, 6(2),
177–190. https://doi.org/10.1080/10447319409526089
Carr, D. F. (2014). Florida Hospital Tracks Nurses Footsteps, Work Patterns. InformationWeek.
https://www.informationweek.com/healthcare/analytics/florida-hospital-tracks-nurses-
footsteps-work-patterns/d/d-id/1127700
Claypoole, V. L., & Szalma, J. L. (2019). Electronic performance monitoring and sustained
attention: Social facilitation for modern applications. Computers in Human Behavior, 94,
25–34. https://doi.org/10.1016/j.chb.2019.01.001
Cohen-Charash, Y., & Spector, P. E. (2001). The role of justice in organizations: A meta-
analysis. Organizational Behavior and Human Decision Processes, 86(2), 278–321.
https://doi.org/10.1006/obhd.2001.2958
Combs, J., Liu, Y., Hall, A., & Ketchen, D. (2006). How much do high-performance work
practices matter? A meta-analysis of their effects on organizational performance. Personnel
Psychology, 59(3), 501–528. https://doi.org/10.1111/j.1744-6570.2006.00045.x
Dahlke, J. A., & Wiernik, B. M. (2019). psychmeta: An R package for psychometric meta-
analysis. Applied Psychological Measurement, 43(5), 415–416.
https://doi.org/10.1177/0146621618795933
DeCaro, M. S., Thomas, R. D., Albert, N. B., & Beilock, S. L. (2011). Choking under pressure:
Multiple routes to skill failure. Journal of Experimental Psychology: General, 140(3), 390–
406. https://doi.org/10.1037/a0023466
Demerouti, E., Nachreiner, F., Bakker, A. B., & Schaufeli, W. B. (2001). The job demands-
A META-ANALYSIS OF THE EFFECTS OF EPM
53
resources model of burnout. Journal of Applied Psychology, 86(3), 499–512.
https://doi.org/10.1037/0021-9010.86.3.499
DeNisi, A. S., & Kluger, A. N. (2000). Feedback effectiveness: Can 360-degree appraisals be
improved? Academy of Management Perspectives, 14(1).
https://doi.org/10.5465/ame.2000.2909845
Doerwald, F., Zacher, H., Van Yperen, N. W., & Scheibe, S. (2021). Generativity at work: A
meta-analysis. Journal of Vocational Behavior, 125, Article 103521.
https://doi.org/10.1016/j.jvb.2020.103521
Dreyfuss, J. (2020). Here’s how employers are using tech tools to keep a close watch on their
remote workers. CNBC. https://www.cnbc.com/2020/06/24/new-tech-tools-employers-are-
using-to-keep-watch-on-remote-workers.html
Dumlao, S. V., Payne, S. C., & George Jr., F. (2019). Employee attitudes towards wearable
technologies in the workplace [Poster presentation]. 34th Annual Conference of the Society
for Industrial and Organizational Psychology.
Earley, P. C. (1988). Computer-generated performance feedback in the magazine-subscription
industry. Organizational Behavior and Human Decision Processes, 41(1), 50–64.
https://doi.org/10.1016/0749-5978(88)90046-5
Eby, L. T., Allen, T. D., Evans, S. C., Ng, T., & DuBois, D. L. (2008). Does mentoring matter?
A multidisciplinary meta-analysis comparing mentored and non-mentored individuals.
Journal of Vocational Behavior, 72(2), 254–267. https://doi.org/10.1016/j.jvb.2007.04.005
Frenkel, S. (2021). A Union Drive at Amazon [Audio podcast episode]. In The Daily. The New
York Times. https://www.nytimes.com/2021/04/01/podcasts/the-daily/amazon-union-vote-
alabama.html?
A META-ANALYSIS OF THE EFFECTS OF EPM
54
Gajendran, R. S., & Harrison, D. A. (2007). The good, the bad, and the unknown about
telecommuting: Meta-analysis of psychological mediators and individual consequences.
Journal of Applied Psychology, 92(6), 1524–1541. https://doi.org/10.1037/0021-
9010.92.6.1524
Geyskens, I., Krishnan, R., Steenkamp, J.-B. E. M., & Cunha, P. V. (2009). A review and
evaluation of meta-analysis practices in management research. Journal of Management,
35(2), 393–419. https://doi.org/10.1177/0149206308328501
Golden, J., & Chemi, E. (2020). Worker monitoring tools see surging growth as companies
adjust to stay-at-home orders. CNBC. https://www.cnbc.com/2020/05/13/employee-
monitoring-tools-see-uptick-as-more-people-work-from-home.html
Green, R. G. (1983). Evaluation apprehension and the social facilitation/inhibition of learning.
Motivation and Emotion, 7(2), 203–212. https://doi.org/10.1007/BF00992903
Greenberg, J. (1987). A taxonomy of organizational justice theories. Academy of Management
Review, 12, 9–22. https://doi.org/10.5465/amr.1987.4306437
Henle, C. A., Kohut, G., & Booth, R. (2009). Designing electronic use policies to enhance
employee perceptions of fairness and to reduce cyberloafing: An empirical test of justice
theory. Computers in Human Behavior, 25(4), 902–910.
https://doi.org/10.1016/j.chb.2009.03.005
Holland, P. J., Cooper, B., & Hecker, R. (2015). Electronic monitoring and surveillance in the
workplace: The effects on trust in management, and the moderating role of occupational
type. Personnel Review, 44(1), 161–175. https://doi.org/10.1108/PR-11-2013-0211
Holman, D., Chissick, C., & Totterdell, P. (2002). The effects of performance monitoring on
emotional labor and well-being in call centers. Motivation and Emotion, 26, 57–81.
A META-ANALYSIS OF THE EFFECTS OF EPM
55
https://doi.org/10.1023/A:1015194108376
Holt, M., Lang, B., & Sutton, S. G. (2017). Potential employees’ ethical perceptions of active
monitoring: The dark side of data analytics. Journal of Information Systems, 31(2), 107–
124. https://doi.org/10.2308/isys-51580
Hovorka-Mead, A. D., Ross, W. H., Whipple, T., & Renchin, M. B. (2002). Watching the
detectives: Seasonal student employee reactions to electronic monitoring with and without
advance notification. Personnel Psychology, 55(2), 329–362.
https://doi.org/10.1111/j.1744-6570.2002.tb00113.x
Hunter, J. E., & Schmidt, F. L. (2004). Methods of meta-analysis: Correcting error and bias in
research findings (2nd ed.). SAGE.
Irving, R. H., Higgins, C. A., & Safayeni, F. R. (1986). Computerized performance monitoring
systems: Use and abuse. Communications of the ACM, 29(8), 794–801.
https://doi.org/10.1145/6424.6430
Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs
and ownership structure. Journal of Financial Economics, 3(4), 305–360.
https://doi.org/10.1016/0304-405X(76)90026-X
Jeske, D., & Kapasi, I. (2017). Electronic performance monitoring: Lessons from the past and
future challenges. In F. Cabitza, A. Lazazzara, M. Magni, & S. Za (Eds.), Organizing for
digital economy: Societies, communities and individuals. Proceedings of the 14th annual
conference of the Italian chapter of the AIS (pp. 119–132). LUISS University Press.
Jeske, D., & Santuzzi, A. M. (2015). Monitoring what and how: Psychological implications of
electronic performance monitoring. New Technology, Work and Employment, 30(1), 62–78.
https://doi.org/10.1111/ntwe.12039
A META-ANALYSIS OF THE EFFECTS OF EPM
56
Kanfer, R., Frese, M., & Johnson, R. E. (2017). Motivation related to work: A century of
progress. Journal of Applied Psychology, 102(3), 338–355.
https://doi.org/10.1037/apl0000133
Karasek, R. A. (1979). Job demands, job decision latitude, and mental strain: Implications for job
redesign. Administrative Science Quarterly, 24(2), 285. https://doi.org/10.2307/2392498
Karim, M. N. (2015). Electronic monitoring and self-regulation: Effects of monitoring purpose
on goal state, feedback perceptions, and learning [Unpublished dissertation, George
Washington University].
https://www.researchgate.net/publication/281102459_Electronic_Monitoring_and_Self-
Regulation_Effects_of_Monitoring_Purpose_on_Goal_State_Feedback_Perceptions_and_L
earning
Katz, D., & Kahn, R. L. (1978). The social psychology of organizations. Wiley.
Kaupins, G., & Coco, M. (2017). Perceptions of Internet-of-Things surveillance by human
resource managers. SAM Advanced Management Journal, 82(2), 53–64.
https://go.gale.com/ps/anonymous?id=GALE%7CA526574031&sid=googleScholar&v=2.1
&it=r&linkaccess=abs&issn=2327087X&p=AONE&sw=w
Landers, R. N., & Marin, S. (2021). Theory and technology in organizational psychology: A
review of technology integration paradigms and their effects on the validity of theory.
Annual Review of Organizational Psychology and Organizational Behavior, 8, 235–258.
https://doi.org/10.1146/annurev-orgpsych-012420-060843
Larson, J. R., & Callahan, C. (1990). Performance monitoring: How it affects work productivity.
Journal of Applied Psychology, 75(5), 530–538. https://doi.org/10.1037/0021-
9010.75.5.530
A META-ANALYSIS OF THE EFFECTS OF EPM
57
Lechermeier, J., & Fassnacht, M. (2018). How do performance feedback characteristics
influence recipients’ reactions? A state-of-the-art review on feedback source, timing, and
valence effects. Management Review Quarterly, 68, 145–193.
https://doi.org/10.1007/s11301-018-0136-8
Levchenko, A. I., Boscart, V. M., & Fernie, G. R. (2011). The feasibility of an automated
monitoring system to improve nurses’ hand hygiene. International Journal of Medical
Informatics, 80(8), 596–603. https://doi.org/10.1016/j.ijmedinf.2011.04.002
Locke, E. A., & Latham, G. P. (2006). New directions in goal-setting theory. Current Directions
in Psychological Science, 15(5), 265–268. https://doi.org/10.1111/j.1467-
8721.2006.00449.x
Lowry, P. B., Posey, C., Bennett, R. J., & Roberts, T. L. (2015). Leveraging fairness and
reactance theories to deter reactive computer abuse following enhanced organisational
information security policies: an empirical study of the influence of counterfactual
reasoning and organisational trust. Information Systems Journal, 25(3), 193–273.
https://doi.org/10.1111/isj.12063
Mamun, M. A. Al, & Yuce, M. R. (2019). Sensors and systems for wearable environmental
monitoring toward IoT-enabled applications: A review. IEEE Sensors Journal, 19(18),
7771–7788. https://doi.org/10.1109/JSEN.2019.2919352
Maskaly, J., Donner, C., Jennings, W. G., Ariel, B., & Sutherland, A. (2017). The effects of
body-worn cameras (BWCs) on police and citizen outcomes: A state-of-the-art review.
Policing, 40(4), 672–688. https://doi.org/10.1108/PIJPSM-03-2017-0032
Mayer, R. C., & Davis, J. H. (1999). The effect of the performance appraisal system on trust for
management: A field quasi-experiment. Journal of Applied Psychology, 84(1), 123–136.
A META-ANALYSIS OF THE EFFECTS OF EPM
58
https://doi.org/10.1037/0021-9010.84.1.123
McDaniel, M. A. (2009). Cumulative meta-analysis as a publication bias method. In 24th Annual
Meeting of the Society for Industrial and Organizational Psychology.
https://pdfs.semanticscholar.org/e125/2957a84e0fb12a86bf086d9a48a0ecd6f079.pdf
McNall, L. A., & Roch, S. G. (2007). Effects of electronic monitoring types on perceptions of
procedural justice, interpersonal justice, and privacy. Journal of Applied Social Psychology,
37(3), 658–682. https://doi.org/10.1111/j.1559-1816.2007.00179.x
McNall, L. A., & Roch, S. G. (2009). A social exchange model of employee reactions to
electronic performance monitoring. Human Performance, 22(3), 204–224.
https://doi.org/10.1080/08959280902970385
McNall, L. A., & Stanton, J. M. (2011). Private eyes are watching you: Reactions to location
sensing technologies. Journal of Business and Psychology, 26, 299–309.
https://doi.org/10.1007/s10869-010-9189-y
Meijman, T. F., & Mulder, G. (1998). Psychological aspects of workload. In P. J. D. Drenth, H.
Thierry, & C. J. de Wolff (Eds.), Handbook of work and organizational psychology (2nd
ed., Vol. 2).
Meluso, J., Johnson, S., & Bagrow, J. (2020). Making virtual teams work: Redesigning virtual
collaboration for the future. https://doi.org/10.31235/osf.io/wehsk
Morris, S., Griffin, D., & Gower, P. (2017, August 18). Barclays puts in sensors to see which
bankers are at their desks. Bloomberg. https://www.bloomberg.com/news/articles/2017-08-
18/barclays-puts-in-sensors-to-see-which-bankers-are-at-their-desks
Nebeker, D. M., & Tatum, B. C. (1993). The Effects of Computer Monitoring, Standards, and
Rewards on Work Performance, Job Satisfaction, and Stress. Journal of Applied Social
A META-ANALYSIS OF THE EFFECTS OF EPM
59
Psychology, 23(7), 508–536. https://doi.org/10.1111/j.1559-1816.1993.tb01101.x
Nishii, L. H., Lepak, D. P., & Schneider, B. (2008). Employee attributions of the “why” of HR
practices: Their effects on employee attitudes and behaviors, and customer satisfaction.
Personnel Psychology, 61(3), 503–545. https://doi.org/10.1111/j.1744-6570.2008.00121.x
Northcraft, G. B., Schmidt, A. M., & Ashford, S. J. (2011). Feedback and the rationing of time
and effort among competing tasks. Journal of Applied Psychology, 96(5), 1076–1086.
https://doi.org/10.1037/a0023221
Posthuma, R. A., Charles Campion, M., & Campion, M. A. (2018). A taxonomic foundation for
evidence-based research on employee performance management. European Journal of
Work and Organizational Psychology, 27(2), 168–187.
https://doi.org/10.1080/1359432X.2018.1438411
Rains, S. A. (2013). The nature of psychological reactance revisited: A meta-analytic review.
Human Communication Research, 39(1), 47–73. https://doi.org/10.1111/j.1468-
2958.2012.01443.x
Ranganathan, A., & Benson, A. (2020). A numbers game: Quantification of work, auto-
gamification and worker productivity.
http://web.stanford.edu/~arunar/docs/ranganathan_benson_2019.pdf
Raveendhran, R., & Fast, N. J. (2021). Humans judge, algorithms nudge: The psychology of
behavior tracking acceptance. Organizational Behavior and Human Decision Processes,
164, 11–26. https://doi.org/10.1016/j.obhdp.2021.01.001
Ravid, D. M., Tomczak, D. L., White, J. C., & Behrend, T. S. (2020). EPM 20/20: A review,
framework, and research agenda for electronic performance monitoring. Journal of
Management, 46(1), 100–126. https://doi.org/10.1177/0149206319869435
A META-ANALYSIS OF THE EFFECTS OF EPM
60
Rogers, S. (2020). Coronavirus has made WFH the new normal. Here’s how virtual reality can
help. Forbes. https://www.forbes.com/sites/solrogers/2020/03/26/coronavirus-has-made-
wfh-the-new-normal-heres-how-virtual-reality-can-help/#4274703f61d5
Rousseau, D. M., Sitkin, S. B., Burt, R. S., & Camerer, C. (1998). Not so different after all: A
cross-discipline view of trust. Academy of Management Review, 23(3), 393–404.
https://doi.org/10.5465/amr.1998.926617
Rudolph, C. W., Katz, I. M., Lavigne, K. N., & Zacher, H. (2017). Job crafting: A meta-analysis
of relationships with individual differences, job characteristics, and work outcomes. Journal
of Vocational Behavior, 102, 112–138. https://doi.org/10.1016/j.jvb.2017.05.008
Salancik, G. R., & Pfeffer, J. (1978). A social information processing approach to job attitudes
and task design. Administrative Science Quarterly, 23(2), 224–253.
https://doi.org/10.2307/2392563
Satariano, A. (2020). How my boss monitors me while I work from home. The New York Times.
https://www.nytimes.com/2020/05/06/technology/employee-monitoring-work-from-home-
virus.html
Schlund, R. J. (2020). I’ll be watching you: AI surveillance technologies and privacy invasion
[Master’s thesis, Cornell University]. https://ecommons.cornell.edu/handle/1813/70277
Schmidt, F. L. (2017). Statistical and measurement pitfalls in the use of meta-regression in meta-
analysis. Career Development International, 22(5), 469–476. https://doi.org/10.1108/CDI-
08-2017-0136
Schmidt, F. L., Hunter, J. E., Pearlman, K., Hirsh, H. R., Sackett, P. R., Schmitt, N., Tenopyr, M.
L., Kehoe, J., & Zedeck, S. (1985). Forty questions about validity generalization and meta-
analysis. Personnel Psychology, 38(4), 697–798. https://doi.org/10.1111/j.1744-
A META-ANALYSIS OF THE EFFECTS OF EPM
61
6570.1985.tb00565.x
Schnackenberg, A. K., & Tomlinson, E. C. (2016). Organizational Transparency: A New
Perspective on Managing Trust in Organization-Stakeholder Relationships. Journal of
Management, 42(7), 1784–1810. https://doi.org/10.1177/0149206314525202
Schoeman, F. D. (1979). On incapacitating the dangerous. American Philosophical Quarterly,
16(1), 27–35. https://doi.org/10.2307/20009736
Schoeman, F. D. (1984). Privacy: Philosophical dimensions. American Philosophical Quarterly,
21(3), 199–213. https://doi.org/10.2307/20014049
Scott, M. B., & Lyman, S. M. (1968). Accounts. American Sociological Review, 33(1), 46–62.
Sherif, K., Jewesimi, O., & El-Masri, M. (2021). Empowering employees: the other side of
electronic performance monitoring. Journal of Information, Communication and Ethics in
Society, 19(2), 207–221.
Silverstein, J. (2018). Walmart patents audio surveillance technology to record customers and
employees. CBS News. https://www.cbsnews.com/news/walmart-patents-audio-
surveillance-technology-to-record-customers-and-employees/
Slemp, G. R., Kern, M. L., Patrick, K. J., & Ryan, R. M. (2018). Leader autonomy support in the
workplace: A meta-analytic review. Motivation and Emotion, 42(5), 706–724.
https://doi.org/10.1007/S11031-018-9698-Y
Smither, J. W., London, M., & Reilly, R. R. (2005). Does performance improve following
multisource feedback? a theoretical model, meta-analysis, and review of empirical findings.
Personnel Psychology, 58(1), 33–66. https://doi.org/10.1111/j.1744-6570.2005.514_1.x
Sonnentag, S., & Frese, M. (2013). Stress in organizations. In N. W. Schmitt, S. Highhouse, & I.
B. Weiner (Eds.), Handbook of psychology: Industrial and organizational psychology (pp.
A META-ANALYSIS OF THE EFFECTS OF EPM
62
560–592). John Wiley & Sons, Inc.
Sonnentag, S., & Fritz, C. (2007). The Recovery Experience Questionnaire: Development and
validation of a measure for assessing recuperation and unwinding from work. Journal of
Occupational Health Psychology, 12(3), 204–221. https://doi.org/10.1037/1076-
8998.12.3.204
Sonnentag, S., Venz, L., & Casper, A. (2017). Advances in recovery research: What have we
learned? What should be done next? Journal of Occupational Health Psychology, 22(3),
365–380. https://doi.org/10.1037/ocp0000079
Sprigg, C. A., & Jackson, P. R. (2006). Call centers as lean service environments: Job-related
strain and the mediating role of work design. Journal of Occupational Health Psychology,
11(2), 197–212. https://doi.org/10.1037/1076-8998.11.2.197
Stanley, T. D., & Doucouliagos, H. (2014). Meta-regression approximations to reduce
publication selection bias. Research Synthesis Methods, 5(1), 60–78.
https://doi.org/10.1002/jrsm.1095
Stanton, J. M. (2000). Reactions to employee performance monitoring: Framework, review, and
research directions. Human Performance, 13(1), 85–113.
https://doi.org/10.1207/S15327043HUP1301_4
Stanton, J. M., & Julian, A. L. (2002). The impact of electronic monitoring on quality and
quantity of performance. Computers in Human Behavior, 18(1), 85–101.
https://doi.org/10.1016/S0747-5632(01)00029-2
Tarafdar, M., Tu, Q., Ragu-Nathan, B. S., & Ragu-Nathan, T. S. (2007). The impact of
technostress on role stress and productivity. Journal of Management Information Systems,
24(1), 301–328. https://doi.org/10.2753/MIS0742-1222240109
A META-ANALYSIS OF THE EFFECTS OF EPM
63
Theeboom, T., Beersma, B., & van Vianen, A. E. M. (2014). Does coaching work? A meta-
analysis on the effects of coaching on individual level outcomes in an organizational
context. The Journal of Positive Psychology, 9(1), 1–18.
https://doi.org/10.1080/17439760.2013.837499
Thompson, L. (2020). Virtual collaboration won’t be the death of creativity. MIT Sloan
Management Review. https://sloanreview.mit.edu/article/virtual-collaboration-wont-be-
%0Athe-death-of-creativity/
U.S. Congress, Office of Technology Assessment. (1987). The Electronic Supervisor: New
Technology, New Tensions.
Valentine, J. C., Pigott, T. D., & Rothstein, H. R. (2010). How many studies do you need? A
primer on statistical power for meta-analysis. Journal of Educational and Behavioral
Statistics, 35(2), 215–247. https://doi.org/10.3102/1076998609346961
Van Iddekinge, C. H., Arnold, J. D., Frieder, R. E., & Roth, P. L. (2019). A meta‐analysis of the
criterion‐related validity of prehire work experience. Personnel Psychology, 72(4), 571–
598. https://doi.org/10.1111/peps.12335
Wang, B., Liu, Y., Qian, J., & Parker, S. K. (2021). Achieving effective remote working during
the COVID‐19 pandemic: A work design perspective. Applied Psychology: An
International Review, 70(1), 16–59. https://doi.org/10.1111/apps.12290
Wells, D. L., Moorman, R. H., & Werner, J. M. (2007). The impact of the perceived purpose of
electronic performance monitoring on an array of attitudinal variables. Human Resource
Development Quarterly, 18(1), 121–138. https://doi.org/10.1002/hrdq.1194
Wilson, T. D., & Gilbert, D. T. (2003). Affective forecasting. Advances in Experimental Social
Psychology, 35, 345–411. https://doi.org/10.1016/S0065-2601(03)01006-2
A META-ANALYSIS OF THE EFFECTS OF EPM
64
Yeginsu, C. (2018, February 1). If Workers Slack Off, the Wristband Will Know. (And Amazon
Has a Patent for It.). The New York Times.
https://www.nytimes.com/2018/02/01/technology/amazon-wristband-tracking-privacy.html
Yost, A. B., Behrend, T. S., Howardson, G., Darrow, J. B., & Jensen, J. M. (2019). Reactance to
electronic surveillance: A test of antecedents and outcomes. Journal of Business and
Psychology, 34(1), 71–86. https://doi.org/10.1007/s10869-018-9532-2
Zajonc, R. B. (1965). Social facilitation. Science, 149(3681), 269–274.
https://doi.org/10.1126/science.149.3681.269
Zeidner, M., & Matthews, G. (2005). Evaluation Anxiety: Current Theory and Research. In A. J.
Elliot & C. S. Dweck (Eds.), Handbook of competence and motivation (pp. 141–163).
Guilford Publications.
Zielinski, D. (2020, August 8). Monitoring Remote Workers. All Things Work.
https://www.shrm.org/hr-today/news/all-things-work/pages/monitoring-remote-
workers.aspx
Zweig, D. (2001). Where is the line between benign and intrusive? An examination of
psychological barriers to the acceptance of awareness monitoring technologies [Doctoral
dissertation, University of Waterloo].
https://search.proquest.com/pqdtglobal/docview/304783634/D516B158A8F04CF1PQ/1?acc
ountid=11243
Zweig, D., & Webster, J. (2002). Where is the line between benign and invasive? An
examination of psychological barriers to the acceptance of awareness monitoring systems.
Journal of Organizational Behavior, 23(5), 605–633. https://doi.org/10.1002/job.157
A META-ANALYSIS OF THE EFFECTS OF EPM
65
Table 1
Summary of Synthetic Construct Groupings for Work Outcomes
Broad Attitudes
Synthetic Construct
Included Variables
Fairness & Justice
Distributive Justice
Interpersonal Justice
Perceived Fairness
Perceived Ethics
Procedural Justice
Satisfaction
Job satisfaction
Policy satisfaction
Task satisfaction
Commitment
Affective Commitment
Continuance Commitment
Normative Commitment
Organizational Commitment
Turnover Intentions
Privacy Invasion
Informational Privacy
Invasion of Privacy
Privacy Concerns
Autonomy
Perceived Autonomy
Perceived Control
Perceived
Support
Perceived Support
Perceived Organizational Support
Monitoring Acceptance
Acceptance of behavior tracking
Monitoring acceptance
Broad Performance
Synthetic Construct Included Variables
Task Performance
Observed Behavioral Performance
Performance Complaints
Performance Ratings
Contextual Performance
Counterproductive Work Behaviors
Organizational Commitment Behaviors
Learning
Post-Training Knowledge
Skill Attainment
Broad Stress/Strain
Synthetic Construct
Included Variables
Stress/Strain
Anxiety
Burnout
Fatigue
Perceived Stress
Physiological Indicators
of Stress (e.g., heartrate)
Pressure
A META-ANALYSIS OF THE EFFECTS OF EPM
66
Table 2
EPM Presence Analysis - Overall Effects of EPM
Work Outcome &
Moderator
K
N
95%CI
80%CR
Broad attitudes
26
5969
-.10
.19
-.11
.21
[-.20, -.03]
[-.37, .14]
Monitoring experiment
12
1812
-.11
.15
-.12
.17
[-.23, -.01]
[-.31, .07]
Vignette experiment
6
1394
-.29
.11
-.32
.10
[-.46, -.19]
[-.48, -.17]
Non-experiment
8
2764
.00
.17
.00
.18
[-.15, .16]
[-.25, .25]
Fairness & justice
9
2390
-.04
.15
-.05
.18
[-.19, .09]
[-.28, .18]
Monitoring experiment
4
1065
.03
.04
.04
.05
[-.04, .11]
[.04, .04]
Non-experiment
3
777
.02
.10
.02
.11
[-.28, -.31]
[-.16, .20]
Satisfaction
14
3552
-.02
.13
-.02
.15
[-.11, .06]
[-.19, .15]
Monitoring experiment
8
993
-.04
.14
-.05
.18
[-.19, .10]
[-.25, .15]
Non-experiment
6
2559
-.01
.12
-.01
.14
[-.16, .13]
[-.21, .18]
Commitment
5
1148
-.03
.20
-.04
.23
[-.33, .26]
[-.38, .31]
Non-experiment
4
1041
.00
.19
.00
.22
[-.35, .35]
[-.34, .34]
Privacy invasion
8
2479
.27
.17
.28
.17
[.13, .43]
[.04, .51]
Monitoring experiment
3
744
.30
.21
.32
.22
[-.23, .87]
[-.09, .72]
Vignette experiment
4
1025
.33
.18
.34
.19
[.04, .63]
[.05, .63]
Autonomy
5
1262
-.15
.22
-.17
.24
[-.46, .13]
[-.52, -.19]
Non-experiment
3
1123
-.11
.08
-.12
.09
[-.33, .10]
[-.24, .01]
Broad performance
41
5804
-.01
.14
-.01
.14
[-.05, .04]
[-.15, .14]
Monitoring experiment
32
3757
-.01
.16
-.01
.17
[-.07, .05]
[-.18, .17]
Non-experiment
7
1816
.00
.09
.00
.09
[-.07, .08]
[-.07, .08]
Task performance
28
3246
.00
.16
.00
.16
[-.06, .06]
[-.16, .17]
Monitoring experiment
26
2972
-.01
.16
-.01
.16
[-.07, .06]
[-.18, .17]
A META-ANALYSIS OF THE EFFECTS OF EPM
67
Speed/quantity
11
1213
-.01
.18
-.01
.18
[-.13, .11]
[-.22, .20]
Accuracy
11
1005
.07
.10
.07
.10
[.00, .14]
[.07, .07]
Performance complaints
4
767
-.06
.05
-.06
.05
[-.13, .02]
[.05, .05]
Contextual performance
9
1922
-.01
.12
-.01
.13
[-.11, .09]
[-.16, .14]
Non-experiment
6
1615
.00
.08
.00
.09
[-.10, .09]
[-.09, .09]
OCBs
4
1012
-.03
.08
-.04
.10
[-.20, .12]
[-.15, .07]
CWBs
6
1575
.00
.11
.00
.13
[-.13, .13]
[-.16, .16]
Learning
4
635
-.02
.11
-.02
.12
[-.21, .16]
[-.16, .11]
Stress/strain
23
5022
.15
.08
.16
.09
[.13, .20]
[.09, .23]
Monitoring experiment
19
2159
.17
.12
.18
.12
[.12, .24]
[.07, .28]
Non-experiment
4
2863
.15
.04
.15
.05
[.08, .23]
[.11, .20]
Physiological stress
4
859
.22
.10
.23
.10
[.06, .40]
[.10, .36]
Note. K = cumulative number of studies; N = cumulative sample size; = sample size-
weighted correlation; = observed standard deviation of ; = sample size-weighted and
reliability-corrected correlation; = standard deviation of ; CI = confidence interval for ;
CR= credibility interval for ; Corrected correlations with confidence intervals that do not
include zero are bolded. Total EPM Presence analysis K = 59, including 23 field studies and 36
laboratory studies.
A META-ANALYSIS OF THE EFFECTS OF EPM
68
Table 3
EPM Presence Analysis with Purpose Included as A Hierarchical Moderator
Purpose
Work Outcome
K
N
95%CI
80%CR
Performance
Broad attitudes
13
2997
-.22
.13
-.23
.15
[-.32, -.14]
[-.41, -.05]
Fairness & justice
6
1646
-.09
.15
-.10
.18
[-.29, .08]
[-.35, .14]
Satisfaction
5
1145
.08
.09
-.10
.10
[-.22, .03]
[-.20, .00]
Privacy Invasion
7
2233
.25
.15
.26
.15
[.12, .40]
[.05, .46]
Broad performance
18
2968
-.03
.14
-.03
.14
[-.10, .04]
[-.18, .13]
Task performance
11
1238
-.04
.18
-.04
.19
[-.16, .09]
[-.26, .18]
Speed/quantity
3
366
-.09
.27
-.09
.27
[-.75, .57]
[-.56, .38]
Accuracy
4
424
.06
.05
.07
.05
[-.01, .14]
[.07, .07]
Contextual performance
5
1294
-.02
.12
-.02
.13
[-.19, .14]
[-.19, .15]
CWB
4
1325
.02
.11
.03
.12
[-.17, .22]
[-.15, .20]
Stress/strain
9
953
.21
.12
.22
.13
[.12, .32]
[.11, .33]
Development
Broad attitudes
5
1102
.03
.20
.04
.23
[-.25, .32]
[-.29, .36]
Admin/Safety
Broad attitudes
5
1442
.01
.04
.01
.05
[-.05, .07]
[.01, .01]
Police BWC
3
569
.03
.05
.03
.06
[-.12, .18]
[.03, .03]
Satisfaction
3
1170
.06
.03
.07
.04
[-.03, .16]
[.07, .07]
Broad performance
13
1528
.04
.13
.05
.14
[-.04, .13]
[-.19, .18]
Police BWC
4
767
.06
.05
.06
.05
[-.02, .14]
[.06, .06]
Without Police BWC
9
761
.03
.19
.03
.20
[-.12, .19]
[-.19, .26]
Task performance
12
1441
.05
.13
.05
.14
[-.03, .14]
[-.08, .19]
Police BWC
4
767
.06
.05
.06
.05
[-.02, .13]
[.06, .06]
Without Police BWC
8
674
.05
.19
.05
.20
[-.12, .21]
[-.18, .28]
Speed/quantity
4
411
.00
.20
.00
.20
[-.32, .32]
[-.29, .28]
Stress/strain
7
1425
.15
.09
.16
.09
[.07, .24]
[.08, .23]
Without Police BWC
6
1154
.15
.10
.16
.10
[.06, .27]
[.06, .26]
Surveillance
Broad attitudes
3
438
-.24
.12
-.26
.13
[-.59, .06]
[-.45, -.08]
A META-ANALYSIS OF THE EFFECTS OF EPM
69
Broad performance
10
971
-.02
.15
-.02
.16
[-.13, .10]
[-.18, .14]
Task performance
8
770
-.01
.14
-.01
.15
[-.14, .11]
[-.16, .13]
Speed/quantity
4
436
.05
.09
.05
.09
[-.09, .20]
[.05, .05]
Accuracy
5
439
.02
.09
.02
.09
[-.10, .13]
[.02, .02]
Stress/strain
6
657
.15
.10
.16
.11
[.05, .28]
[.11, .22]
Note. K = cumulative number of studies; N = cumulative sample size; = sample size-weighted
correlation; = observed standard deviation of ; = sample size-weighted and reliability-
corrected correlation; = standard deviation of ; CI = confidence interval for ; CR =
credibility interval for ; Corrected correlations with confidence intervals that do not include
zero are bolded. Six cross-sectional studies that aggregated information across individuals who
were monitored in many types of ways and provided no information about purpose were coded
as “presence” and included in the broad presence analysis, but not in any analysis of EPM with a
specific purpose.
A META-ANALYSIS OF THE EFFECTS OF EPM
70
Table 4
EPM Presence Analysis with Invasiveness Included as a Hierarchical Moderator
Invasiveness
Work Outcome
K
N
95%CIL
80%CR
Task-target
Broad attitudes
14
3828
-.06
.19
-.07
.22
[-.19, .06]
[-.34, .21]
Fairness & justice
3
719
-.01
.13
-.01
.15
[-.37, .35]
[-.25, .23]
Satisfaction
7
1378
-.04
.13
-.05
.15
[-.18, .09]
[-.23, .14]
Commitment
3
355
-.25
.08
-.29
.09
[-.52, -.06]
[-.29, -.29]
Privacy Invasion
4
1683
.19
.12
.19
.12
[.00, .39]
[.01, .38]
Broad performance
25
3502
-.01
.11
-.01
.12
[-.05, .04]
[-.10, .09]
Task performance
17
1667
.00
.13
.00
.13
[-.07, .06]
[-.11, .10]
Speed/quantity
8
941
-.01
.17
-.01
.17
[-.15, .14]
[-.21, .20]
Accuracy
9
863
.04
.07
.04
.07
[-.02, .10]
[.04, .04]
Contextual performance
5
1342
.00
.09
.00
.10
[-.12, .13]
[-.11, .12]
CWB
3
608
-.05
.08
-.06
.09
[-.28, .16]
[-.13, .01]
Stress/strain
14
1942
.17
.12
.19
.13
[.11, .26]
[.07, .30]
Person-target
Broad attitudes
14
2979
-.13
.23
-.14
.25
[-.29, .00]
[-.47, .18]
Fairness & justice
6
1565
-.09
.16
-.10
.19
[-.30, .10]
[-.36, .16]
Satisfaction
5
1176
.05
.15
.05
.18
[-.17, .28]
[-.19, .30]
Privacy Invasion
5
1041
.44
.10
.46
.10
[.33, .58]
[.33, .58]
Broad performance
17
2123
.00
.19
.00
.20
[-.10, .10]
[-.23, .23]
Task performance
13
1729
.01
.18
.01
.19
[-.10, .12]
[-.21, .23]
Speed/quantity
3
272
-.02
.26
-.02
.26
[-.66, .63]
[-.46, .43]
Contextual performance
3
307
-.05
.25
-.06
.29
[-.79, .66]
[-.57, .44]
Stress/strain
10
1665
.18
.11
.19
.11
[.10, .27]
[.07, .30]
Note. K = cumulative number of studies; N = cumulative sample size; = sample size-
weighted correlation; = observed standard deviation of ; = sample size-weighted and
reliability-corrected correlation; = standard deviation of ; CI = confidence interval for ;
A META-ANALYSIS OF THE EFFECTS OF EPM
71
CR= credibility interval for ; Corrected correlations with confidence intervals that do not
include zero are bolded. Task-targeted is EPM that strictly focuses on task related information
(e.g., typing speed); Person-targeted is EPM that targets a person’s body, location, or thoughts
and feelings (e.g., email-content monitoring or GPS tracking). Thoughts, feelings, and
physiology were initially coded separate from person, but later combined with person due to a
very small sample (K =3).
Table 5
EPM Presence Analysis with Synchronicity Included as a Hierarchical Moderator
Synchronicity
Work Outcome
K
N
95%CI
80%CR
Synchronous
Collection
Broad attitudes
13
2033
-.26
.12
-.29
.13
[-.37, -.21]
[-.43, -.16]
Fairness & justice
3
675
-.22
.18
-.20
.12
[-.76, .25]
[-.60, .10]
Satisfaction
5
461
-.18
.07
-.21
.09
[-.32, -.10]
[-.21, -.21]
Privacy Invasion
5
1129
.33
.17
.35
.17
[.13, .57]
[.10, .60]
Broad performance
25
2388
-.03
.18
-.04
.19
[-.11, .04]
[-.24, .17]
Task performance
21
1940
-.02
.19
-.02
.19
[-.11, .06]
[-.23, .19]
Speed/quantity
10
1074
-.01
.19
-.01
.19
[-.15, .13]
[-.24, .22]
Accuracy
11
1005
.07
.10
.07
.10
[.00, .14]
[.07, .07]
Stress/strain
16
1457
.18
.13
.20
.14
[.12, .27]
[.09, .31]
Asynchronous
Collection
Broad attitudes
8
2080
.00
.21
.00
.23
[-.19, .19]
[-.31, .31]
Fairness & justice
4
1138
.04
.04
.04
.05
[-.03, .12]
[.04, .04]
Satisfaction
5
1241
.04
.16
.05
.18
[-.17, .28]
[-.20, .30]
Commitment
3
833
.03
.20
.04
.24
[-.55, .63]
[-.39, .47]
Broad performance
13
2102
.01
.11
.01
.12
[-.06, .08]
[-.11, .13]
Task performance
8
1349
.03
.09
.03
.09
[-.05, .11]
[-.04, .10]
A META-ANALYSIS OF THE EFFECTS OF EPM
72
Contextual performance
3
320
-.05
.24
-.06
.28
[-.75, .64]
[-.54, .43]
Stress/strain
7
1500
.17
.07
.18
.08
[.11, .25]
[.13, .23]
Note. K = cumulative number of studies; N = cumulative sample size; = sample size-
weighted correlation; = observed standard deviation of ; = sample size-weighted and
reliability-corrected correlation; = standard deviation of ; CI = confidence interval for ;
CR = credibility interval for ; Corrected correlations with confidence intervals that do not
include zero are bolded. Synchronous Collection is EPM in which performance information is
collected, analyzed, and evaluated in real-time (e.g., a supervisor watching a live feed of
employee performance); Asynchronous Collection is EPM in which performance information is
recorded or archived to be viewed or analyzed later. Synchronicity of feedback was also coded
for but was not included in results due to insufficient sample variance (i.e., K< 3 studies coded as
synchronous).
Table 6
EPM Degree Analysis
EPM
Characteristic
Work Outcome
K
N
95%CI
80%CR
Invasiveness
Broad attitudes
30
10904
-.15
.12
-.17
.13
[-.22, -.11]
[-.32, -.01]
Monitoring experiment
4
530
-.20
.19
-.22
.21
[-.55, .11]
[-.52, .08]
Vignette experiment
11
5109
-.12
.12
-.13
.14
[-.22, -.04]
[-.30, .04]
Non-experiment
15
5265
-.17
.11
-.19
.12
[-.26, -.13]
[-.34, -.05]
Fairness & justice
11
3766
-.12
.15
-.14
.17
[-.25, -.02]
[-.35, .08]
Monitoring experiment
3
422
-.22
.12
-.24
.12
[-.55, .07]
[-.40, -.08]
Vignette experiment
4
2145
-.06
.08
-.06
.09
[-.21, .08]
[-.19, .06]
Non-experiment
4
1199
-.21
.22
-.23
.24
[-.61, .15]
[-.61, .15]
Satisfaction
4
703
-.05
.23
-.05
.25
[-.45, .35]
[-.44, .33]
Commitment
4
1004
-.15
.09
-.18
.11
[-.36, -.01]
[-.30, -.07]
Non-experiment
3
628
-.14
.12
-.18
.14
[-.54, .18]
[-.39, .04]
Privacy invasion
13
4652
.18
.15
.20
.16
[.10, .29]
[-.01, .40]
A META-ANALYSIS OF THE EFFECTS OF EPM
73
Vignette experiment
6
2605
.12
.13
.13
.14
[-.02, .28]
[-.06, .32]
Non-experiment
5
1704
.25
.17
.27
.18
[.04, .49]
[.00, .53]
Autonomy
11
3695
-.15
.11
-.18
.13
[-.26, -.09]
[-.33, -.02]
Non-experiment
8
3304
-.14
.12
-.17
.14
[-.28, -.05]
[-.34, .01]
Monitoring acceptance
4
2882
-.19
.14
-.20
.15
[-.44, .04]
[-.43, .04]
Broad performance
9
2451
-.11
.14
-.13
.16
[-.25, -.01]
[-.32, .07]
Non-experiment
7
2266
-.10
.14
-.12
.16
[-.26, .03]
[-.32, .09]
Task performance
4
1044
-.05
.13
-.05
.15
[-.28, .18]
[-.26, .15]
Contextual performance
5
1407
-.16
.14
-.18
.16
[-.38, .01]
[-.40, .04]
CWBs
4
1217
.20
.12
.22
.14
[.00, .44]
[.02, .42]
Stress/strain
10
3260
.17
.11
.20
.13
[.10, .29]
[.04, .35]
Non-experiment
7
2785
.19
.09
.21
.10
[.12, .30]
[.10, .33]
Burnout
3
1591
.15
.04
.18
.05
[.06, .30]
[.18, .18]
Breadth
Broad attitudes
18
7077
-.16
.12
-.18
.14
[-.25, -.11]
[-.35, -.01]
Vignette experiment
4
2021
-.08
.17
-.09
.19
[-.39, .21]
[-.39, .20]
Non-experiment
13
4850
-.18
.08
-.21
.09
[-.26, -.16]
[-.29, -.12]
Fairness & justice
7
3117
-.01
.11
-.01
.12
[-.12, .10]
[-.16, .14]
Vignette experiment
3
1937
.03
.06
.04
.07
[-.13, .21]
[-.06, .14]
Non-experiment
3
974
-.04
.08
-.05
.08
[-.25, .16]
[-.15, .06]
Privacy Invasion
10
3931
.21
.19
.23
.20
[.09, .37]
[-.03, .50]
Vignette experiment
4
2021
.07
.14
.08
.15
[-.16, .31]
[-.15, .31]
Non-experiment
5
1704
.35
.05
.38
.05
[.31, .44]
[.38, .38]
Autonomy
7
3114
-.15
.12
-.18
.14
[-.30, -.05]
[-.36, .00]
Broad performance
6
2076
-.11
.14
-.13
.16
[-.30, .05]
[-.35, .10]
Contextual performance
4
1217
-.19
.13
-.21
.14
[-.44, .02]
[-.42, .00]
CWBs
4
1217
.20
.12
.22
.14
[.00, .44]
[.02, .42]
Stress/strain
7
2785
.19
.09
.22
.10
[.12, .31]
[.10, .33]
Burnout
3
1591
.15
.04
.18
.05
[.06, .30]
[.18, .18]
Constraints/
Target
Control
Broad attitudes
16
6794
.15
.15
.17
.16
[.08, .25]
[-.03, .37]
Monitoring experiment
3
393
.12
.14
.13
.15
[-.24, .49]
[-.08, .34]
Vignette experiment
9
4364
.12
.11
.13
.12
[.04, .22]
[-.02, .28]
Non-experiment
4
1037
.33
.18
.36
.20
[.04, .68]
[.05, .67]
Fairness & Justice
7
2292
.25
.17
.27
.19
[.10, .45]
[.02, .53]
Vignette experiment
3
1484
.16
.05
.18
.05
[.05, .30]
[.18, .18]
Satisfaction
3
377
-.06
.15
-.07
.16
[-.46, .33]
[-.30, .17]
Privacy invasion
7
2688
-.19
.12
-.20
.13
[-.32, -.08]
[-.36, -.04]
Vignette experiment
4
1860
-.14
.10
-.15
.10
[-.32, .02]
[-.30, .00]
Autonomy
4
581
.20
.05
.22
.06
[.12, .31]
[.22, .22]
Monitoring acceptance
3
2221
.13
.06
.13
.07
[-.04, .29]
[.02, .23]
A META-ANALYSIS OF THE EFFECTS OF EPM
74
Broad performance
3
375
.13
.16
.14
.16
[-.27, .54]
[-.12, .39]
Transparency
Broad attitudes
14
3696
.22
.22
.25
.24
[.11, .39]
[-.07, .57]
Monitoring experiment
3
334
.02
.29
.02
.32
[-.78, .81]
[-.55, .59]
Vignette experiment
5
1941
.11
.03
.13
.03
[.08, .17]
[.13, .13]
Non-experiment
6
1421
.42
.21
.47
.24
[.22, .72]
[.13, .80]
Fairness & justice
9
2632
.19
.20
.21
.22
[.04, .38]
[-.09, .51]
Vignette experiment
5
1941
.11
.06
.12
.06
[.04, .20]
[.07, .17]
Satisfaction
8
1745
.23
.19
.26
.21
[.08, .44]
[-.02, .54]
Non-experiment
5
1213
.29
.17
.34
.19
[.10, .58]
[.06, .62]
Commitment
5
1164
.29
.17
.35
.21
[.10, .61]
[.06, .65]
Affective commitment
3
501
.19
.13
.22
.15
[-.16, .59]
[-.01, .45]
Privacy invasion
4
1608
-.06
.07
-.07
.08
[-.19, .06]
[-.16, .03]
Vignette-experiment
3
1471
-.08
.04
-.08
.05
[-.20, .03]
[-.08, -.08]
Perceived Support
5
1143
.32
.17
.37
.20
[.13, .62]
[.09, .66]
Non-experiment
3
871
.38
.12
.45
.14
[.10, .80]
[.21, .68]
Broad performance
6
1001
.08
.09
.09
.11
[-.03, .21]
[.00, .18]
Non-experiment
4
758
.08
.11
.10
.14
[-.12, .31]
[-.07, .26]
Task Performance
3
500
.05
.02
.06
.03
[.00, .12]
[.06, .06]
Contextual performance
3
501
.10
.14
.13
.17
[-.30, .56]
[-.14, .40]
Note. K = cumulative number of studies; N = cumulative sample size; = sample size-weighted
correlation; = observed standard deviation of ; = sample size-weighted and reliability-
corrected correlation; = standard deviation of ; CI = confidence interval for ; CR=
credibility interval for ; Corrected correlations with confidence intervals that do not include
zero are bolded; Invasiveness = EPM that is greater in breadth, higher in specificity, more
personal in target, and has fewer constraints and less target control. EPM constraints and target
control were initially coded separately, but later combined into a single variable due to
conceptual similarity between the two characteristics. Analysis with and without studies coded as
constraints revealed no significant differences in effects. EPM Degree analysis K = 44, including
25 field studies and 19 laboratory studies.
A META-ANALYSIS OF THE EFFECTS OF EPM
75
Table 7
Avenues for Future EPM Research and Example Research Questions
Unstudied and Understudied Areas From the EPM Characteristics Typology
• Does Development Purpose moderate the effects of EPM on attitudes, performance, and
stress/strain?
• Under what circumstances do individuals find safety-focused EPM acceptable/unacceptable?
• Does synchronicity of feedback moderate the effects of EPM on work outcomes (e.g.,
performance, stress)?
• How do individuals respond to highly invasive forms of EPM (e.g., physiological monitoring,
social media monitoring)?
• Does transparency in monitoring moderate the effects of EPM on performance outcomes?
Purpose as Perceptions
• Do specific profiles of purpose perceptions (e.g., high in Admin/Safety and high in
Development) moderate the effects of EPM on attitudes, performance, and stress/strain
• How do purpose perceptions develop, change, strengthen, or weaken with time and
experience?
Boundary Conditions and Mediators
• Do individual differences (e.g., intelligence, neuroticism, trait reactance, age, experience with
monitoring) interact with the characteristics of monitoring to predict work outcomes?
• To what extent does work setting (e.g., central office, home office) moderate the effects of EPM
on work outcomes such as privacy invasion and perceived justice?
• Do occupational characteristics (e.g., job complexity, automatability) interact with EPM
characteristics to influence individual-level work outcomes?
• Does national culture (e.g., uncertainty avoidance) interact with EPM characteristics to influence
individual-level work outcomes?
• Do the characteristics of monitoring interact with the passage of time to influence work
outcomes (i.e., do individuals desensitize or sensitize to the effects of EPM over time)?
• Do more proximal outcomes of EPM (e.g., privacy invasion, loss of autonomy, justice
perceptions) mediate the relationships between EPM characteristics and other work outcomes?
Unexamined Potential Outcomes of EPM
• What is the effects of safety focused EPM on safety behaviors and safety climate?
• Does the presence of EPM technologies during non-work periods (e.g., rest breaks) influence
recovery experiences (e.g., detachment)?
• Does the presence of EPM influence individual- and team-level creativity?
• Does EPM influence organization-level work outcomes such as work climate and culture?
• Does EPM affect individuals’ perceptions of management and organizational leadership?
A META-ANALYSIS OF THE EFFECTS OF EPM
76
Appendix
Table A1
Typology of EPM Characteristics with Monitoring Examples
EPM Element
Sub-Element
Category
Examples in Practice
Purpose
The explicit or
perceived rationale
for EPM use
Performance appraisal,
loss prevention, and
profit
• Webpage browsing is tracked via
computer monitoring software to deter
off task behaviors
• Product assembly speed is tracked via
smart cameras (e.g., Amazon Panorama
cameras) with metrics used to evaluate
performance
Development, growth,
and training
• Employees are provided real time
computerized feedback during a
Microsoft Excel training
• Welders are provided with automatized
welding cues to help improve their skills
Administrative and
safety
• A therapist logs session notes in
electronic health systems for liability
purposes
• Location tracking at construction site to
alert individuals if they have stepped
into an unsafe zone
Surveillance or
authoritarian
• An employer uses computer software to
collect employee data (e.g., how long
word documents are open) that have no
clear link to performance criteria
• Installing monitoring (e.g., cameras)
technologies without articulating any
rationale
Invasiveness
Scope
Breadth
The amount, target,
and systematic
constraints placed on
EPM use
High
• A wide variety of employee work
behaviors at a call center are tracked
(e.g., call time, call content, employee
affect, number and length of breaks)
Low
• A very narrow and limited set of
university professor work behaviors
(e.g., classroom lectures) are recorded
Specificity
High
• Performance behaviors of individual
manufacturing employees are tracked
and evaluated
Low
• A retailer monitors sales at the group
level (e.g., sales of the clothing
department)
Target
Thoughts, feelings, and
physiology
• Biometric sensors monitor fatigue in
heavy-machine operators
• Employer monitoring of employee Slack
exchanges
Person and location
• Location tracked RFID nurse badges
• Video surveillance of an office space
A META-ANALYSIS OF THE EFFECTS OF EPM
77
Task
• A grocery store tracks how many
customers each cashier checks out.
Constraints
High
• EPM data are stored securely and can
only be accessed for safety purposes by
top management
Low
• EPM data are housed in a public
database can be used for a variety of
purposes
Target Control
High
• A worker can turn off internet activity
tracking to use Facebook
Low
• A rideshare driver has no control to
disable location tracking while on the
job
Synchronicity
The temporal
aspects of EPM use
including frequency
and regularity of
monitoring
Collection
High
• Theft prevention actively and
continuously monitors employee
behaviors at a department store via
security cameras
Low
• Screen shots of employee work
computers are intermittently captured
and stored for viewing at a later time
Feedback
High
• A surgeon wears smart glasses and
receives real-time feedback during a
procedure
Low
• A recruiter receives a monthly report of
the number and average duration of calls
made
Transparency
The extent to which
employees are
provided
information about
the characteristics of
monitoring
High
• A company explicitly outlines EPM
practices in employee handbooks;
managers discuss EPM practices face-to-
face with new hires
Low
• A company does not outline in writing
or formally discuss EPM practices;
employees speculate what EPM data are
used for
