Content uploaded by Carey Doberstein
Author content
All content in this area was uploaded by Carey Doberstein on Aug 04, 2020
Content may be subject to copyright.
An Empirical Assessment of the Intrusiveness and Reasonableness of
Emerging Work Surveillance Technologies in the Public Sector
Étienne Charbonneau
École nationale d’administration publique
4750 Henri-Julien, 5th floor
Montreal, QC
Canada
H2T 3E5
etienne.charbonneau@enap.ca
https://orcid.org/0000-0003-0194-2983
Carey Doberstein
Department of Political Science
University of British Columbia
C425 1866 Main Mall
Vancouver, BC
Canada
V6T 1Z1
carey.doberstein@ubc.ca
https://orcid.org/0000-0001-5106-8484
Étienne Charbonneau is Canada Research Chair in Comparative Public Management at
École nationale d’administration publique in Montreal. His recent research delves on
accountability and electronic surveillance.
Carey Doberstein is assistant professor at the department of Political Science at the
University of British Columbia in Vancouver. He is an associate editor of Canadian
Public Administration. His latest book is “Distributed Democracy: Health Care
Governance in Ontario,” with University of Toronto Press.
ACKNOWLEDGEMENTS
We thank the members of the Canadian Public Sector Research Panel for participating in
the research. We also thank Pr Nicholas Jobidon for his initial suggestions about legal
jurisprudence, and Itizez Slama for her work with the panel's recruitment.
This article is protected by copyright. All rights reserved.
Accepted Article
This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process which may lead to
differences between this version and the Version of Record. Please cite this article as doi:
10.1111/puar.13278
Abstract: As public sector work environments continue to embrace the digital
governance revolution, questions of work surveillance practices and its relationship to
performance management continue to evolve, but even more dramatically in the
contemporary period of many public servants being forced to shift to remote working
from home in response to the COVID-19 pandemic. This article presents the results of
three surveys, two of them population-based survey experiments, all conducted during
the onset of the COVID-19 pandemic in Canada that compare public servant (n= 346)
and citizen (n=1,008 phone; n=2,001 web) attitudes to various cutting-edge—though no
doubt controversial among some— digital surveillance tools that can be used in the
public sector to monitor employee work patterns, often targeted towards remote working
conditions. The findings represent data that can help governments and public service
associations navigate difficult questions of reasonable privacy intrusions in an increasing
digitally-connected workforce.
Evidence for Practice
New work surveillance technologies are available to use within the public sector and will
present acceptability challenges to public managers as they contemplate their
introduction.
Multi-modal survey data from Canada reveals that public servants and citizens find these
emerging work surveillance technologies to be quite intrusive and unreasonable, but
show relatively more tolerance for digital surveillance over physical surveillance
practices.
Understanding surveillance anxieties among targeted employees will be key to finding a
balance between employee privacy rights and employer desires to manage employees in
remote or digital environment.
Introduction
This article is protected by copyright. All rights reserved.
Accepted Article
In the midst of the COVID-19 pandemic, social workers in the Canadian province
of Quebec were initially forced by their public sector managers to continue their onsite
visits of needy families and in-person team meetings, violating the physical distancing
protocols from the Government of Quebec. Managers were apparently concerned with
maintaining productivity and chipping away at the waitlists even in a crisis context, afraid
that remote working would compromise these efforts (Allard 2020). The Minister later
denounced the managers and demanded remote working during this period, consistent
with the advice of public health experts on the necessary steps to stem the spread of the
virus. The COVID-19 pandemic dramatically accelerated an existing trend towards
remote work in the public sector, introducing uncertainty with respect to the security of
data, the management of the work of teams, as well as individual productivity, in an
environment when nearly entire organizations are working virtually from home. The
COVID-19 pandemic has ushered in a period, the length of which is unknown, in which
remote working is the norm, and public sector work environments forced to change quite
dramatically while under incredible pressure to serve the government and citizens. With
so many public sector employees working from home, this raises unique supervisory
demands on human resources managers (Schuster et al. 2020, p.3), in particular manager-
employee relationships, work assignments and accountability. One critical piece of this
equation relates to work surveillance, and the technological advancements that have made
digital work surveillance in the context of remote working not only more possible, but
also potentially more fraught with privacy concerns.
Work surveillance can take various forms, from light observation to extremely
intrusive observation, from digital surveillance to more complex artificial intelligence
This article is protected by copyright. All rights reserved.
Accepted Article
analysis. As of June 2020, an uptick in the use of surveillance software was documented
for some private firms (Thompson 2020), but not for public organizations in Canada and
abroad. A piece of software gaining in popularity with private sector employers since the
COVID-19 pandemic is an always-on video conference call software that takes pictures
every few minutes via a front-facing laptop webcam, and posts them on a wall for
managers to see them working at their desks (Holmes 2020). There is also software and
hardware for monitoring devices (e.g. telephone, email, texts, keystrokes, clicks on
computer/internet), one’s environment (e.g. geospatial movements, desk sensors at
workspace) and via biometrics (e.g. heart rate, facial recognition) that are now easily
integrated into most workplaces. Even more complex and potentially intrusive
surveillance is apparent in police departments pre-COVID-19. To fulfill a mandate from
a federal judge, the police department in Oakland, California uses an artificial
intelligence software to flag possible cases of policy officer corruption by compiling and
analyzing “all aspects of an officer’s career, including time in the academy, data on
police stops, citizen complaints, body camera footage and use of force” (Cassidy 2019).
Likewise, the Massachusetts State Police in 2020 will introduce geolocalization
technology in their 2,900 vehicles to track and validate officer movements throughout
shifts (Miller 2020).
Other examples of workplace surveillance can similarly bring clearly into view
the privacy implications of the increasingly easy, but also invasive, monitoring potential
of employers. In 2019, a class-action lawsuit emerged in Illinois on the use of
biometrics—in this case fingerprint technology—against employers of Thyssenkrupp
Crankshaft, United Airlines and Hilton Hotels and Resorts for using fingerprint scanners
This article is protected by copyright. All rights reserved.
Accepted Article
to prevent “buddy punching” (co-workers punching in a friend who is late or absent from
work). The plaintiffs argued in part that this is an unreasonable infringement on their
privacy to be forced to submit biometric data to the employer. What is considered a
reasonable expectation of privacy in any given society or context is not fixed in time
(Kugler 2014, 1206-1207), and is especially subject to evolution as new technologies
become embedded in our personal and professional lives. For some observers, as new
technologies enter the market and workplaces, “…our expectations of privacy diminish
from what is reasonable to what is merely foreseeable” (Johnson 2012, 415). Something
analogous to “behavioral fatigue,” as is speculated with social distancing during the
current pandemic (Pedersen and Favero 2020, 12), might be at play, as emotionally and
cognitively depleted employees, do not resist to new demands put on them.
As consumers, individuals are perhaps growing accustomed to being surveilled by
free services offered by Google and Facebook, or paid Microsoft software, like Cortana,
or premium devices like the Apple Watch. It is unclear if individuals, as workers, have
higher expectations of privacy. The digital revolution was already transforming
traditional forms of surveillance—swinging by office desk, secret shoppers, sign out
boards—into potentially much more continuous, passive, and all-encompassing
surveillance in a context whereby so much of work exists in cyber space (Ajunwa et al.
2017). This type of digital surveillance falls under the umbrella of electronic performance
monitoring (EPM)—even if the purpose is not strictly for performance evaluation or
enhancements—for which there is enormous diversity in terms of its frequency, purpose,
scope, and transparency to the employee (Ravid et al. 2020). The COVID-19 pandemic
has accelerated this trend, and thus foists upon public sector work environments
This article is protected by copyright. All rights reserved.
Accepted Article
important questions about what digital work surveillance practices are reasonable
intrusions on the privacy of employees.
There is complex legal terrain to navigate in Anglo-American contexts with
respect to privacy, informed by broad constitutional provisions, employment law, and
even collective agreements, all of which are interpreted by human resource managers,
professional associations, arbitrators, and judges to reach decisions about what constitutes
invasions of privacy by public organizations. For example, in Canada, case law
recognizes elements of privacy (Khullar 2012, 392), even if other arbitrators or legal
actors consider this a “legal fiction or a misunderstanding” (Khullar 2012, 385). The U.S.
and Canada fall into the same group of common law and statutory rights of action
involving privacy torts, as opposed to the UK and Australia where breaches of confidence
are the main issue (Paterson 2018, 209). And among various Anglo-American legal
regimes, there is enough commonality that courts occasionally draw on case law from
other countries, as was the case in Canada, where the Supreme Court recognized
arguments from the early privacy case of Katz v. United States (1967) to provide clarity
on the privacy dimensions of Canada’s Charter of Rights and Freedoms (Johnson 2012,
472).
Yet such court decisions of what is a reasonable expectation of privacy are not
solely normative or deduced from first principles. Chao and colleagues (2018, 265) argue
that “(…) all too often those assertions are based not on reliable empirical data, but rather
on judges’ intuitions and sheer speculations.” This is buttressed by experimental survey
research nearly 30 years ago by Slobogin and Schumacher (1993), who compared the
expectations of privacy of the public with judges’ views expressed in key court decisions
This article is protected by copyright. All rights reserved.
Accepted Article
(regarding police behaviors), and found that judges misjudged what the public found
intrusive or reasonable in terms of police searches, seizures and surveillance (Slobogin
and Schumacher 1993, 774). Since then, a number of similar studies have emerged
(including Blumenthal et al. 2009, Fradella et al. 2011, Kugler 2014, Scott-Hyward et al.
2015, Smith et al. 2016, McAllister 2018, Chao et al. 2018) as part of the Empirical
Legal Studies movement (Ho and Rubin 2011). This type of empirical research has not
been done in a Canadian context, and to our knowledge, not focused on professional
public servants elsewhere. There is thus not only an opportunity to settle to what degree
public sector work privacy is a contested public value (Bozeman 2019), but also to
broaden this kind of inquiry beyond criminal and police enforcement contexts, as Hoetger
(2013) contends that the findings from police searches vis-à-vis privacy “would likely
extend to employer searches” (581).
This research provides an empirical assessment of what public servants and
citizens in Canada find reasonable and intrusive in terms of electronic workplace
surveillance in the public sector, as this data can inform, in part, the foundation of the
“reasonableness” standard in human resource management policy and legal tests. As we
will describe in our methods section, our two citizen-surveys (web and phone) are
population-based survey experiments which aim to maximize internal and external
validity. While our methodology mimics Slobogin and Schumacher’s (1993) and others
who followed, our two large representative samples of citizens permit us to offer a
reliable snapshot to practitioners, compared to previous studies with a small number of
participants. A 2014 study of senior government officials involved in national security
This article is protected by copyright. All rights reserved.
Accepted Article
decision making revealed that representative polls are the most desirable quantitative
studies by practitioners (Avey and Desch 2014, 231-232).
This study thus provides guidance to human resource managers calibrating their
work surveillance policies and practices, as well as produces empirical data for the
inevitable future legal proceedings that grapple with tests of reasonableness, which
heretofore emerge primarily from legal reasoning and a weak empirical grounding in
Canada (Geist 2003, 178) and the United States (Ciochetti 2011, 356). Geist (2003, 178)
explains that Canadian and U.S. jurisprudence has been shifting slowly from only
considering whether there is a reasonable expectation of privacy in workplace contexts,
to an “emerging analysis [that] focuses instead on whether the surveillance itself is
reasonable”, which is an area in which social scientists can contribute.
The article proceeds as follows. First, we review the literature that speaks to
workplace surveillance, electronic performance management and emerging technologies
in the public sector, and following that review the comparative legal terrain for privacy in
the workplace in Anglo-American contexts. Second, we describe the methodology of this
study, which gathers survey data from a panel of Canadian public servants, as well as the
broader public, to provide an empirical foundation for tests of the reasonableness of
twelve existing and cutting-edge work surveillance technologies that are emerging (or are
likely be proposed soon) in the public sector. Following that we present the data and
analysis of the survey results, revealing that public servants and citizens similarly find
these emerging work surveillance technologies to be quite intrusive and unreasonable, but
both show greater tolerance for digital surveillance over physical surveillance practices.
The final section explores the implications of these findings for public sector
This article is protected by copyright. All rights reserved.
Accepted Article
environments as governments and public service associations navigate difficult questions
of reasonable privacy intrusions in an increasing digitally-connected workforce.
LITERATURE REVIEW
Electronic performance monitoring (EPM) in workplaces
A starting point for the study of work surveillance can be traced to Taylor (1912)
as part of scientific management approaches focused on efficiency and performance of
workers and workflows, objectives that can be tracked much more easily in modern
digital work environments than in Taylor’s time (Ciochetti 2011, 285-286). Electronic
performance monitoring (EPM), broadly speaking, is enabled by the omnipresence of
digital devices in modern organizations, as well as work patterns such as an increasing
number of professional public sector employees who work remotely (Fusi and Feeney
2018). Whereas Taylor was limited to the performance managers could surveil in person,
and with considerable human hours devoted to do so, EPM approaches can be
continuous, discreet, intrusive, conducted without warning or consent, and in many cases
can be analyzed automatically (Ravid et al. 2020). Ajunwa et al. (2017) note a shift from
what they call the more traditional “authoritarian surveillance” that was imposed from
above to the more recent phenomenon of “participatory surveillance” whereby employees
are asked or expected to use apps or digital workspaces that purport to be beneficial to
them (e.g. wellness apps, productivity apps, etc.).
Types of EPM used in modern organizations, both in the private and public sector
include device monitoring (e.g. telephone email, texts, keystrokes, clicks on
computer/internet), environmental monitoring (e.g. cameras, geospatial movements, desk
This article is protected by copyright. All rights reserved.
Accepted Article
sensors at workspace, handwashing badges), and biometric monitoring (e.g. wellness
apps, heart rate, facial recognition). Each of these types can be implemented such that the
scope of data collection is broad or alternatively narrow in scope, as well as continuously
or intermittently collected, and for basic accounting (e.g. minutes at desk) or more
interpretive analysis (e.g. tone of keywords in emails to estimate office mood).
Depending on the context, work surveillance may be aimed towards data-driven
performance measurement, reducing or eliminating personal tasks during work hours
(e.g. coordinating day care pick up, online banking, etc.), or regulating remote working
conditions (Ankabi 2017).
EPM has been studied since the 1980s, and whereas early studies were
inconclusive on the relationship between the presence of EPM and performance or
employee attitudes, contemporary research has discovered that there are many
contingencies involved in this equation, in particular that employee effects from EPM
depend on a interaction between the purpose, target, intensity, scope and feedback
mechanisms involved in the surveillance, as well as individual-level attributes of the
employee, such as their trust in management (Ravid et al. 2020). Furthermore, the same
technology can be at times seen as caring or coercive, depending on the motivations
perceived by workers (Anteby and Chan 2018, 248). Why this all matters is that when
employees perceive unfair monitoring, they are more likely to report lower job
satisfaction and greater stress (Young 2010) and even engage in resistance and creative
avoidance (Kayas et al. 2019).
Workplace privacy rights in Anglo-American contexts
This article is protected by copyright. All rights reserved.
Accepted Article
The pervasiveness of EPM in modern workplaces—public and private sector—
presents critical perennial and new questions of privacy rights for employees. Balancing
privacy rights of employees with employer desires to surveil their environments has long
been a struggle in public law and human resources management, but the surveillance
possibilities with new technologies presents unique challenges to this equilibrium. The
core foundational principle in Anglo-American law regarding privacy at the workplace
revolves around the concept of “reasonableness” (Fric 2016, 63-65), for which there are
several dimensions: Is there a reasonable expectation of privacy in this setting? Was the
purpose of the search by the employer reasonable? Was the search conducted in a
reasonable manner? While not all tests established by courts in Anglo-American contexts
are the same, they are similarly grounded in the discourse of “reasonableness” (Hoetger
2013; Fric 2016; Hunt and Bell 2015).
American jurisprudence has loomed large in legal framing and thinking on work
surveillance in Canada (Geist 2003, 163; Eltis 2006, 479), though Canada has preserved
the expectation of privacy at work to a larger degree (Phillips 2015, 479). Yet courts and
arbitrators in the United States and Canada have “no bright-line rule for evaluating
government employee’s expectations for privacy” (Hoetger 2013, 568). For example, for
camera-based work surveillance in Canada, one of the few technologies where a rich
jurisprudence exists, Khullar (2012, 383-389) tracked down five tests, with seventeen
questions, ten mentions of reasonableness and two explicit mentions of intrusiveness.
Most of those tests involve answering questions about whether the surveillance is
demonstrably necessary to meet a specific need, the likelihood of its effectiveness in
meeting that need, a proportionality test of loss of privacy vis-à-vis benefit gained, and
This article is protected by copyright. All rights reserved.
Accepted Article
confirmation that there are no less invasive way to achieve the same end (Levin 2007).
Yet answering the questions in these tests ultimately come down to the subjective
opinions of judges and arbitrators.
One case exemplifies the apparent subjectivity of legal tests in a Canadian context:
Erwin Eastmond v. Canadian Pacific Railway and Privacy Commissioner of Canada
(2004). Different courts, using different tests of what is intrusive and reasonable, arrived
at different verdicts of what a reasonable person may conclude as the same case snaked
up to higher courts. Blasina (2007, 468) concludes from the long Eastmond legal saga
that “the circumstances may allow different adjudicators to come to different conclusions
of fact, although they have given reasonable consideration to the same body of evidence.”
With so much of the analysis of the balance between the privacy of employees and the
rights of employers to observe their work environment hinging on questions of
“reasonableness”, it is important to explore how that standard is determined. Smith et al.
(2016) argue that U.S. courts have tended to not attempt to assess views of
reasonableness from the public itself, but rather from their own sense of reasonableness;
this is echoed by Eltis (2015) with respect to Canadian courts.
While most case law in Anglo-American contexts has moved beyond a simple
assessment of an employee’s reasonable expectation of privacy in particular work
environments—and instead moving towards assessing as well whether the surveillance
itself is reasonable on a host of criteria—this does not address the problem of courts and
arbitrators relying primarily on their subjective sense of reasonableness. Reasonableness
as a concept ought not to be deduced solely from logic but also from the views of citizens
in this context, given that we know expectations and demands for privacy are continually
This article is protected by copyright. All rights reserved.
Accepted Article
evolving. Yet there are risks in viewing reasonableness by a societal standard informed
from aggregated public opinion, as that may be shaped by the very technological
advancements that are threatening privacy at work. In particular, Eltis argues
“if privacy continues to be defined by reference to reasonable expectations,
technological imperatives necessarily dictate that the sphere in which one can
reasonably claim solitude will decrease. In other words, assessing an individual’s
right to privacy by reference to society’s conception of the measure of privacy
that one is entitled to reasonably expect is particularly awkward when such
expectations are rapidly eroding, precisely by reason of eventual social
habituation to recurring intrusions” (Eltis 2015, 496).
Having a baseline of employee and citizen approval before a technology is widespread is
thus useful. Likewise, having an empirical basis for statements about reasonableness in
the context of work surveillance technologies can complement logic-based and
contextualized tests established by courts and arbitrators in the pursuit of less-arbitrary
findings.
Empirically Measuring Reasonableness and Intrusiveness
Remarking on American judges’ perception of what citizens find reasonable when
interacting with the police, Wilson (2008, 40) argues that jury members would be better
suited than judges to have a sense of what the public thinks. Ascertaining what society
finds reasonable in terms of work surveillance is an empirical question with an empirical
(though not fixed in time) answer. Slobogin and Schumacher (1993) discovered in
surveys that members of the American public found many practices deemed “reasonable”
by American judges about law enforcement as invasive and in violation of their
expectations of privacy. The authors used online experimental vignettes to test the effect
of a first/third person view and the absence/presence of evidence on the respondents’
This article is protected by copyright. All rights reserved.
Accepted Article
view on intrusiveness of police searches. They surveyed a small sample of students—
some in law schools, some not—on their views of surveillance activities with the aim to
establish relative intrusiveness among different practices.
Blumenthal et al. (2009) essentially replicated Slobogin and Schumacher’s (1993)
study with 158 undergraduate psychology students, mirroring their findings, but also
noting that those students who were exposed to more contextual information about the
surveillance were less likely to find it intrusive than those asked about a particular
surveillance practice in the abstract. Fradella et al. (2011) surveyed students and faculty
at eleven universities, as well as the public via invitations from Facebook, on privacy as it
relates to the body, territory, information and communications, and their agreement with
thirty-five judicial decisions in this realm. They concluded that “collectively, the results
indicate that courts often misjudge what ‘society’ is prepared to embrace as a reasonable
expectation of privacy” (372).
Scott-Hayward et al. (2015) likewise find that the courts are often widely out of
step with what the public expects in terms of privacy on five common practices of the
third-party doctrine (which enables police to acquire information on online activities and
localizations from social media companies, internet and phone providers, without
probable cause). The most spectacular gap between the courts’ views of what the public
expects and what respondents expect is that:
Almost 90% of participants felt that law enforcement should never have access, or
at least require a level commensurate with probable cause to obtain information
about email addresses with which an individual has been in contact. This is in stark
contrast to the leading circuit court decision, holding that users have no expectation
of privacy in this information. (Scott-Hayward et al. 2015, 54)
This article is protected by copyright. All rights reserved.
Accepted Article
Chao et al. (2018) also surveyed 1,200 U.S. respondents to compare their
expectations of privacy to the ones depicted in court decisions and similarly found that
judges overestimate the public’s tolerance for police surveillance. The discrepancies for
digital surveillance are especially important:
(…) survey participants consider five of the technology searches to be the most
intrusive of all the study’s scenarios. All five of these—Stingray devices, drones,
obtaining emails, accessing the Cloud, and GPS tracking—were considered more
intrusive than a police search of one’s bedroom, the quintessential violation of
privacy that requires justification by probable cause and a warrant. (Chao et al.
2018, 309-310)
In light of empirical legal studies such as these, Scott-Hayward et al. (2015)
suggest that judges should consult robust public opinion data and not lean exclusively on
their own sensitivities about what appears reasonable to citizens (Scott-Hayward et al.
2015, 58). While all of the studies cited above are focused on police surveillance, the
questions of reasonableness and intrusiveness of surveillance are parallel to those in the
context of work surveillance in the public sector. From these kinds of studies, we are able
to identify where the majority of respondents land on the technologies and thus derive
where the “reasonable person” stands. Furthermore, the findings related to the enhanced
feelings of intrusiveness of digital surveillance are particularly important to explore in the
context of work surveillance in the digital era, and especially so in context of the
COVID-19 pandemic when much of the public service is working remotely and will be
for some time.
Surveillance technologies in the workplace
This article is protected by copyright. All rights reserved.
Accepted Article
While there are conceptual parallels in police surveillance cases in Anglo-
American contexts with workplace surveillance in the public sector, the technologies
used and their contexts are quite different. Thus, while serving as the inspiration for this
study, the relationship between a citizen and the state (ie. police surveillance) is different
from that of an employee and employer (ie. work surveillance), and as a result we must
compile an empirical evidence base specific to public sector work surveillance and
questions of reasonableness and intrusiveness.
Writing from a Canadian context, Levin (2007, 216-217) provided a list of
electronic surveillance methods used or piloted by employers ranging from CCTVs,
biometric identifiers, radio frequency identifiers (RFIDs), internet and email monitoring,
and keystroke tracking. Since that time of writing, we know of additional surveillance
practices used in the public and private sector around the world. For example, a device
called “OccupEye” was used at The Telegraph (UK) newspaper, which is a box placed
under one’s desk that can track attendance and take body heat measures as part of a
productivity assessment of employees (Ajunwa et al. 2017). At Deloitte and Bank of
America, employees have worn “Humanyze badges” which can see and hear everything
they do, and analyze their speech volume and pitch, track who they spend time with and
the physical path of their day (Steele 2020). While keyloggers have been a surveillance
option for employers for two decades (Geist 2003), newer software such as Clickstream
can collect very specific data and produce reports on how people use their computer and
internet throughout the day (Ajunwa et al. 2017). Remote workers may also be digitally
surveilled with increased ease, with examples from ODesk.com (now called UpWork)
whereby a photo is taken by front-facing camera randomly six times every hour to
This article is protected by copyright. All rights reserved.
Accepted Article
estimate time at one’s desk (Ajunwa et al. 2017) and at Crossover Worksmart whereby
software takes screenshots at random to produce a “digital timecard” every ten minutes to
determine the amount of paid time working (Captain 2020). While the private sector is
where many of these technologies have been piloted, they are increasingly present in the
public sector as well, as described explicitly in the new “Policy on Service and Digital”
by the Government of Canada (Government of Canada 2020). Some surveillance tools
have clearer performance purposes compared to some other surveillance without clear
purposes. There are many more examples of cutting edge surveillance technologies in
workplaces in Anglo-American contexts in addition to those identified above that we test
explicitly in our survey of Canadians, the design of which is detailed in the next section.
STUDY DESIGN
‘Bureaucrat’ is an amorphous term. The dimension we focus on is surveillance for
two vivid examples of public servants with (typically) higher and lower appreciation
among the public, a social worker and government tax agent, respectively (Fukuyama
2013; Tummers et al. 2013). The empirical strategy for this study is to survey Canadians
on questions of the reasonableness and intrusiveness of a dozen digital surveillance
measures for two types of public sector workers (randomly assigned to respondents): (i) a
public servant in the national capital working for the tax collection agency, and (ii) a
social worker in an organization funded by government. These two areas of public
service provide us with a maximal range of likability, which might impact one’s
assessment. There is one survey, administered to different subpopulations and completed
both online and by phone. The first subpopulation is a sample of Canadian public
servants from the Canadian Public Sector Research (CPSR) Panel created by the authors,
This article is protected by copyright. All rights reserved.
Accepted Article
which is a voluntary panel of public servants willing to answer online academic surveys
related to the work they do for Canadians. At the time the survey was conducted from
March 11th-26th, 2020, there were 1,206 members, 346 out of 402 of whom completed the
surveys with few missing data, for a response rate of 27%. The second subpopulation of
Canadians surveyed online was a sample of 2,001 Canadian citizens with the help of
Léger Marketing, a national polling company, surveyed from March 17-26th, 2020. In this
sample, we intentionally oversampled (n= 1,001) young adults (aged 18-30) in order
evaluate hypotheses that younger citizens have lower expectations of privacy and higher
degrees of comfort with technologies (Chao et al. 2018), as part of an analytical blocking
strategy (Mutz 2011). As such, the sample of Canadians (n= 1,000) collected by Léger in
the online survey was a nationally-representative sample, aside from age.
The final subpopulation is also a representative sample of another 1,008
Canadians, but it was administered by phone by Mainstreet Research, a national polling
company. The motivation behind using a bi-modal data collection strategy for the
population-based citizen samples is to mitigate bias of the survey method, which was
observed by Boivin and Cordeau (2017) in their study of citizen satisfaction with the
Montreal Police Department, finding that telephone respondents were more likely to
express satisfaction than online respondents even after controlling for demographics,
victimization, and lifestyle. This was also discovered by Herian and Tomkins (2012) in a
similar study in Nebraska. Furthermore, other researchers have speculated that members
of the public who are registered with an online market research panel may have particular
sensibilities towards digital surveillance than the wider public (Chao et al. 2018;
Abraham et al. 2019) that we are able to test (and control for in our analysis).
This article is protected by copyright. All rights reserved.
Accepted Article
The central task asked of all survey respondents was for them to identify, in their
opinion, the “reasonableness” and “intrusiveness” of twelve emerging workplace
surveillance measures for public sector employees. Similar to Slobogin and Schumacher
(1993) who use experimental vignettes to examine the impact of contextual elements on
perceptions of their respondents, we have included technologies with clear performance
purposes as well as those without clear purposes, as well as randomly vary the type of
public sector worker for whom these surveillance measures are targeting: a public servant
in Ottawa working for the Canada Revenue Agency or a social worker in an organization
funded by government.
Just as Slobogin and Schumacher (1993) used experimental vignettes to study the
impact of contextual elements on the perceptions of their respondents, we borrow the
same strategy in this study. Table 1 below summarizes the items tested for reasonableness
and intrusiveness. In terms of measures of reasonableness and intrusiveness, we followed
methods used by Kugler (2014, 1194), Chao and colleagues (2018, 295) and Slobogin
and Schumacher (1993, 736) who use a 0-100 intrusiveness scale, and a six-point
reasonableness scale.
[table 1 about here]
This survey is limited to Canadian public servants and citizens, but we believe the
empirical story has the potential to inform debates in a cross-national context. To provide
evidence of the extent of comparability of our Canadian sample to an American context,
we replicate a survey question advanced by Rainie and Duggan (2016) at the Pew
This article is protected by copyright. All rights reserved.
Accepted Article
Research Center to Americans (n=461) in relating to work surveillance. The question is
as follows, and was replicated in our survey:
Several co-workers of yours have recently had personal belongings stolen
from your workplace, and the company is planning to install high-resolution
security cameras that use facial recognition technology to help identify the
thieves and make the workplace more secure. The footage would stay on file
as long as the company wishes to retain it, and could be used to track various measures
of employee attendance and performance.
Would the scenario be acceptable to you, or not? (Yes, No, It depends (please explain))
To the extent that Canadian respondents mirror the results from this question, we will be
able to speak with greater confidence of the cross-national implications of our findings.
We also asked all respondents questions about their views of their own colleagues (Yin et
al. 2013) at work (or previous job if currently unemployed) and their general trust in
others, as potential intervening variables on their reported perceptions of reasonableness
and intrusiveness of various workplace surveillance technologies. The full survey can
viewed in Appendix A1, as well as a demographic summary of all respondents, public
servants and the representative samples of Canadian citizens.
RESULTS
On the measure of intrusiveness of the twelve work surveillance technologies, we
observe a distinct pattern vis-à-vis public servant respondents and citizen respondents, as
shown below in Table 2: generally a similar order with respect to the technology’s
intrusiveness, but shifted approximately 10-17 points towards more intrusiveness
according to public servant respondents compared to citizen respondents. This is
expected given that the question is focused on the surveillance of a public sector worker,
and thus public servants are likely thinking of the intrusiveness of these technologies in a
This article is protected by copyright. All rights reserved.
Accepted Article
very personal manner, whereas citizens are thinking about surveilling others—a key
distinction observed in the literature (Slobogin and Schumacher 1993). Note that we
inspected the results for any patterns in intrusiveness and reasonableness scores with
respect to gender and ethnicity and found no large independent effects of these
dimensions (age is explored further below).
One finding of particular note, however, is that computer software surveillance
(e.g. internet usage, Clickstream software, key logger, AI email analysis) is viewed as
slightly less intrusive than cameras, photo capture and Humanyze badges (which is
basically a personal audio recorder’ of one’s work day), despite actually gathering more
information on an individual. This may be due to camera and camera-like surveillance
being perceived as surveillance without purpose or related meaningfully to performance,
but rather as a blunt surveillance instrument for general observance. Other arguably more
invasive technologies are more tolerated likely because respondents can see their
relationship to work productivity or performance, in particular that the surveillance
produces a desirable effect on worker and team behavior.
[table 2 about here]
There is a very strong correlation between one’s sense of the intrusiveness of a
technology and their views of its reasonableness for use in a public sector work
environment. There are surveillance technologies outside of the realm of work, however,
that are generally viewed as quite intrusive but nonetheless may be viewed as reasonable
to use in certain contexts (e.g. ankle monitors as conditional for parole). For the work
This article is protected by copyright. All rights reserved.
Accepted Article
surveillance technologies examined here, keycards with RFID, which can capture
movements around the office environment, was in the low range of intrusiveness
according to our respondents (including public servants), but was viewed as unreasonable
by public servants. This is likely due to an absence of an obvious surveillance-
performance association, which can contribute to the rejection of some types of intrusive
technologies.
Previous research suggests that a lack a reciprocal trust between political officials
and bureaucrats contributes to more technological surveillance within the bureaucracy
(Fusi and Feeney 2018, 1470), so to what extent does this extend to citizens vis-à-vis
public servants? In contrast to Fusi and Feeney (2018), our samples included members of
the public, who mostly work in the private sector, although some are public servants
themselves. We were also interested in examining how trust in one’s colleagues, and
others more broadly, may shape one’s sense of the intrusiveness and reasonableness of
various work surveillance technologies. Individuals who bear low trust for their fellow
coworkers might very well be more supportive of technologies surveilling their activities.
The effect of individuals with lower trust in most people or distrust in supervisors on
support for work surveillance has not been firmly established in the literature (Nakhai
and de Lint 2013; Weckert 2002). Figure 1 reveals how both one’s trust in colleagues, as
well others more generally, moderately contributes to intrusiveness and reasonable scores
across all the technologies (aggregated), differentiated by the two types of public sector
workers who would be targeted by the work surveillance: a public servant in Ottawa
working for the Canada Revenue Agency or a social worker in an organization funded by
government.
This article is protected by copyright. All rights reserved.
Accepted Article
[figure 1 around here]
The essential takeaway from Figure 1 is that we find no systematic patterns
among our respondents with regard to the intrusiveness or reasonableness of a work
surveillance technology that is conditional on the type of public sector worker, and that
while trust in colleagues and trust in others shapes the aggregated scores of intrusiveness
and reasonableness, the substantive differences are negligible. The survey mode (online
versus phone), however, seemingly produced much larger effects on intrusiveness and
reasonable scores, with those contacted by phone (n= 1,008) tending to find work
surveillance technologies more intrusive and less reasonable than the Canadians
contacted via a web panel (n=2,001). We believe this to be a combination of the
purposeful over-sampling of younger cohorts in the web panel, as well as the mode (ie.
participants on web panels are likely more comfortable with technology). All that said,
when trust in colleagues, trust in others and the survey mode are held constant, the
differences between public sector workers with higher and lower appreciation—social
workers and Canada Revenue agents, respectively—are not large enough to be
statistically significant.
[figure 2 around here]
To further investigate the potential of an age cohort effect, Figure 2 illustrates the
differences between younger (18-30, n=1,169) and older (30+, n=1,840) citizens in terms
of their intrusiveness and reasonableness scores for twelve work surveillance
This article is protected by copyright. All rights reserved.
Accepted Article
technologies (representative web panel and weighted-representative phone survey data
pooled). Younger citizens find wellness apps and handwashing badges on balance not
intrusive and most lean towards a position of them being "somewhat reasonable."
Younger citizens mostly agree with other citizens as to the intrusiveness and
reasonableness of OccupyEye, AI email software, and key loggers, even if they find them
slightly less objectionable. Nevertheless, younger citizens find random photo capture
software and internet usage monitoring statistically more intrusive and less reasonable
than older cohorts. Age is not a distinguishing factor for the intrusiveness and
reasonableness scores for non-visible cameras, Humanyze badges, Clickstream software,
facial recognition, and RFID.
The final area of data analysis speaks to the comparability of the findings from a
Canadian context to elsewhere, in particular to the United States. Recall that we
replicated a question from Rainie and Duggan’s (2016) study (n=461) with regard to the
acceptability of office surveillance cameras with facial recognition capability for the
purposes of security and performance analysis. Our large Canadian sample from the web
panel mirrors closely the findings from the smaller U.S. online panel respondents on this
question, with 52.3% finding them acceptable (z=0.67; n.s.), compared to 54%, However,
our representative phone respondents differ widely from the U.S. web panel participants
28.8% finding it acceptable (z=9.31; p<.001). This may be explained by the fact that
while both self-selecting to participate in a study, web panel participants are different
than phone respondents, as they had to register to a panel prior to a study, hence self-
selecting twice. Phone surveys are costly, more than ten dollars per respondent, rather
than 55 cents per respondents on crowdsourcing platforms (Pedersen and Favero 2020,
This article is protected by copyright. All rights reserved.
Accepted Article
p.27), but they reach citizens who are seldom studied otherwise, in particular those who
may exhibit reticence with regards to the internet. Regardless, the close alignment of our
web panel sample across the two countries to the same question related to work
surveillance lends confidence to drawing inferences from this study to the American
context.
DISCUSSION
In designing this study, we set out to find—largely through media stories and tech
journalism—the most cutting edge workplace surveillance technologies being proposed
or used in Anglo-American contexts, and take a temperature on them among the
Canadian public and public servants with regard to their sense of the technology’s
intrusiveness. We discovered that the Canadian public and public servants find most of
the emerging surveillance technologies quite intrusive and their use unreasonable in
public sector work environments. Yet as we have discovered earlier in the review of court
and arbitration cases, judges and arbitrators are, by contrast, much more permissive of
camera and phone technologies in the workplace, and thus operate on assumptions about
reasonableness that are often disconnected from the public. Determining the balance of
worker privacy rights and employer desire to know the happenings of the workplace,
particularly in the context of a sudden (and likely sustained) surge in remote working as a
result of COVID-19, is key going forward.
From this study, we are able to deduce a number of dimensions of privacy
concerns among those who would be surveilled, as well as the broader expectations of the
citizenry with respect to what is reasonable, as depicted in Table 3 below. Technologies
This article is protected by copyright. All rights reserved.
Accepted Article
that aim to capture the physical activity of employees (cameras, movement trackers, etc.)
are viewed as slightly more intrusive than digital footprint technologies (keystroke
loggers, Clickstream software, email AI analysis), despite the latter actually collecting
more useful information from a productivity standpoint. All of the ‘very unreasonable’
work surveillance technologies in Table 3 have in common a lack of a clear surveillance-
performance link; in other words, the surveillance is not so much associated with their
work productivity and task achievements, but about monitoring them or their bodies in
the environment. It is also worth being mindful of the age cohort effects of work
surveillance attitudes in the context of recruitment and retention in the public sector; new
recruits are likely to be especially repulsed by work surveillance technologies like
random photo capture software and internet usage monitoring.
[Table 3 here]
This speaks to the emotional content of workplace surveillance technologies that
can influence morale and anxiety, and thus knowing the nature of the concerns of types of
work surveillance on the employee side, and citizens more generally, as well as tolerance
for some types of surveillance, is important for public sector employers, human resource
managers, as well as arbitrators and courts in the cases of disputes. An empirical basis for
claims of intrusiveness and reasonableness is an essential supplement to deductively
derived principles of legal concepts like ‘reasonableness’, and can aide in achieving a
more systematic balance, and avoiding the arbitrary and conflicting findings we see in
This article is protected by copyright. All rights reserved.
Accepted Article
Anglo-American contexts with regard to reasonable and unreasonable workplace
surveillance (Fradella et al. 2011, 371).
In their study, Slobogin and Schumacher (1993, 743) proffered that justices of the
U.S. Supreme Court could use their results on work surveillance in four different ways:
reject or ignore them, change their legal analysis to make the results of the judges-citizen
gap null, incorporate the results in the way they reach decisions (including reversing
themselves), or finally build a new way to model their decisions in the future to reflect
the views of the members of the society they serve. Yet most of the questions about
public sector work surveillance will not be sorted out in the courts, but in the choices of
public managers and human resource teams in conjunction with public service unions and
privacy commissioners, which Fusi and Fenney (2018) confirm with their finding of the
enormous variation in the use of work surveillance technologies in the U.S. This research
is aimed at building an empirical base from which to inform managers and policymakers’
ongoing conversation about appropriate public sector work surveillance that truly
balances worker privacy rights and concerns with the employer’s desire to maintain a
productive workplace in a context of growing remote working outside of the traditional
office environment. By focusing on the technologies that are certainly soon to be
proposed, we are able to provide public sector parties a foundational analysis of these
emerging technologies in relation to one another, as well as a framework in which to
examine the trade-offs before they are put in place and become the source of controversy.
While we believe that an empirical basis for discussions is an important
complement to negotiations over public sector work surveillance, there are limitations to
this research, though much of which can be addressed in future research. First, when
This article is protected by copyright. All rights reserved.
Accepted Article
workplace privacy cases are before arbitrators or judges, they are considering many more
elements and contextual details of a particular dispute; the item-vignette technologies in
our survey are stripped of such context and thus they represent abstractions, not cases. As
others before us acknowledged (Chao et al. 2018, 316), that is an inherent limitation to
this method. At the same time, we could not have surveyed the public expecting them to
read the amount of material that judges consider, but perhaps richer (but fewer) vignettes
that add more contextual dimensions could be illuminating in a future study. A second
limitation is that the Canadian sample may not travel as far as we assume. While we
attempt to address this with a replication question from a 2016 Pew study of Americans
to establish a measure of alignment of samples, further studies will need to investigate for
the presence of country-specific norms of public sector work surveillance.
CONCLUSION
In their essay on the ethics of work surveillance in the public sector, West and
Bowman (2016, 637) stated that “privacy is not just an individual right but also a societal
good: The presumption of freedom and independence from being constantly watched and
the ability to create one’s professional role in an authentic manner. Strong reasons, then,
must exist to subject employees to surveillance.” We agree with this sentiment, and
further argue that we have to be very careful about work surveillance as the technological
capacity advances in such a way that it allows it to be done in a much more continuous,
passive, and all-encompassing manner. Described as the third shock to the millennial
paradigm, the COVID 19 pandemic has the potential to introduce a measure presented as
temporary, but that can become the new normal (Roberts 2020, 1). The ability to do it—
This article is protected by copyright. All rights reserved.
Accepted Article
and the relatively low-cost options from companies that provide such surveillance
services—does not mean public sector employers should do it or that it is the best manner
in which to promote high performance among employees. Yet at the same time we ought
to appreciate that the workplace is changing dramatically as part of a longer-term trend,
with surges in remote work outside the traditional environment in the context of COVID-
19 and certainly thereafter, in which public managers need to responsibly adapt. The
private sector is pioneering the use of these surveillance technologies—often with push
back from employees—but the public sector in particular needs to get this right, as they
are working on behalf of citizens to support the governments they elect, and thus we are
all invested in a public sector work environment that achieves a mutually acceptable—
and reasonable—balance of privacy and surveillance in these settings.
REFERENCES
Abraham, Martin, Niessen, Cornelia, Schnabel, Claus, Lorek, Kerstin, Grimm, Veronika,
Möslein, Kathrin, & Wrede, Matthias. (2019). Electronic Monitoring at Work: The Role
of Attitudes, Functions, and Perceived Control for the Acceptance of Tracking
Technologies. Human Resource Management Journal, 29(4), 657-675.
Ajunwa, Ifeoma, Crawford, Kate, & Schultz, Jason. (2017). Limitless Worker
Surveillance. California Law Review, 105(3), 735-776.
Allard, Marc. (2020, March 27th). COVID-19: «Les travailleurs sociaux n’ont pas tous
besoin d’être sur le terrain», dit le ministre Carmant. Le Soleil. Quebec City, QC.
Anteby, Michel, & Chan, Curtis K. (2018). A Self-Fulfilling Cycle of Coercive
Surveillance: Workers’ Invisibility Practices and Managerial Justification. Organization
Science, 29(2), 247-263.
Avey, Paul C, & Desch, Michael C. (2014). What Do Policymakers Want from Us:
Results of a Survey of Current and Former Senior National Security Decision Makers.
International Studies Quarterly, 58(2), 227-246.
This article is protected by copyright. All rights reserved.
Accepted Article
Blumenthal, Jeremy A., Adya, Meera, & Mogle, Jacqueline. (2009). The Multiple
Dimensions of Privacy: Testing Lay "Expectations of Privacy". Journal of Constitutional
Law, 11(2), 331-373.
Boivin, Rémi, & Cordeau, Gilbert. (2017). Do Web Surveys Facilitate Reporting less
Favourable Opinions about Law Enforcement? Security Journal, 30(2), 335-348.
Bozeman, Barry. (2019). Public Values: Citizens’ Perspective. Public Management
Review, 21(6), 817-838.
Captain, Sean. (2020, January 27th). In 20 Years, Your Boss May Track Your
Every Glance, Keystroke, and Heartbeat. Boston, MA: Fast Company.
Cassidy, Megan. (2019, November 25th). Oakland, Calif., to Use Tech to Fight Police
Corruption. San Francisco, CA: San Francisco Chronicle.
Chao, Bernard, Durso, Catherine, Farrell, Ian, & Robertson, Christopher. (2018). Why
Courts Fail to Protect Privacy: Race, Age, Bias, and Technology. California Law Review,
106(2), 263-324.
Eltis, Karen. (2005). La surveillance du courrier électronique en milieu de travail: le
Québec succombera t-il à l’influence de l'approche américaine. McGill Law Journal,
51(3), 475-502.
Eltis, Karen. (2015). Piecing Together Jones, A.B. and Cole: Towards a “Proportional”
Model of Shared Accountability in Workplace Privacy. Canadian Labour and
Employment Law Journal, 18(2), 493-515.
Fric, Agathon. (2016). Reasonableness as Proportionality: Towards Better Constructive
Interpretation of the Law on Searching Computers in Canada. Appeal: Review of Current
Law and Law Reform, 21, 59-82.
Fradella, Henry F., Morrow, Weston J., Fischer, Ryan G., & Ireland, Connie. (2011).
Quantifying Katz: Empirically Measuring Reasonable Expectations of Privacy in the
Fourth Amendment Context. American Journal of Criminal Law, 38(3), 289-374.
Fukuyama, Francis. (2013). What is Governance? Governance, 26(3), 347-368.
Fusi, Frederica, & Feeney, Mary K. (2018). Electronic Monitoring in Public
Organizations: Evidence from U.S. Local Governments. Public Management Review,
20(10), 1465-1489.
Geist, Michael A. (2003). Computer and E-Mail Workplace Surveillance in Canada: The
Shift from Reasonable Expectation of Privacy to Reasonable Surveillance. Canadian Bar
Review, 82(2), 151-190.
This article is protected by copyright. All rights reserved.
Accepted Article
Government of Canada (2020). Policy on Service and Digital. Ottawa, ON: Treasury
Board of Canada. https://www.tbs-sct.gc.ca/pol/doc-eng.aspx?id=32603. Accessed on
April 3, 2020.
Herian, Mitchel N., & Tomkins, Alan J. (2012). Citizen Satisfaction Survey Data: A
Mode Comparison of the Derived Importance-Performance Approach. American Review
of Public Administration, 42(1), 66-86.
Ho, Daniel E., & Rubin, Donald B. (2011). Credible Causal Inference for Empirical
Legal Studies. Annual Review of Law and Social Science, 7, 17-40.
Hoetger, Lori. (2013). Did My Boss Just Read That? Applying a Coding vs. Content
Distinction in Determining Government Employees’ Reasonable Expectation of Privacy
in Employer- Provided Electronic Communication Devices After City of Ontario v.
Quon, 130 S. Ct. 2619 (2010). Nebraska Law Review, 90(2), 559-585.
Holmes, Aaron. (2020, March 23rd). Employees at Home are Being Photographed every 5
Minutes by an Always-on Video Service to Ensure they're Actually Working — And the
Service is Seeing a Rapid Expansion since the Coronavirus Outbreak. Business Insider.
New York, NY.
Hunt, Chris, & Bell, Corinn. (2015). Employer Monitoring of Employee Online
Activities outside the Workplace: Not Taking Privacy Seriously. Canadian Labour and
Employment Law, 18(2), 411-458.
Johnson, Mathew. (2012). Privacy in the Balance - Novel Search Technologies,
Reasonable Expectations, and Recalibrating Section 8. Criminal Law Quarterly, 58(3),
442-509.
Kayas, O. G., Hines, T., McLean, R., & Wright, G. H. (2019). Resisting Government
rendered Surveillance in a Local Authority. Public Management Review, 21(8), 1170-
1190.
Kugler, Matthew B. (2014). The Perceived Intrusiveness of Searching Electronic Devices
at the Border: An Empirical Study. University of Chicago Law Review, 81(3), 1165-1211.
Khullar, Ritu. (2012). Influence of Oakes Outside the Charter, Specifically Labour
Arbitration Jurisprudence. Ottawa Law Review, 43(3), 377-393.
Levin, Avner. (2007). Big and Little Brother: The Potential Erosion of Workplace
Privacy in Canada. Canadian Journal of Law and Society, 22(2), 197-230.
McAllister, Marc. (2014). GPS and Cell Phone Tracking: A Constitutional and Empirical
Analysis. University of Cincinnati Law Review, 82(1), 207-256.
This article is protected by copyright. All rights reserved.
Accepted Article
Miller, Norman. (2020, January 10th). Massachusetts State Police Adopt New Officer-
Tracking Tech. Framingham, MA: Metrowest Daily News.
Mutz, Diana C. (2011). Population-Based Survey Experiments. Princeton, NJ: Princeton
University Press.
Nakhaie, Reza, & de Lint, Willem. (2013). Trust and Support for Surveillance Policies in
Canadian and American Opinion. International Criminal Justice Review, 23(2), 149-169.
Paterson, Moira. (2018). Regulating Surveillance: Suggestions for a Possible Way
Forward. Canadian Journal of Comparative and Contemporary Law, 4(1), 193-230.
Pedersen, Mogens Jin, & Favero, Nathan. (2020). Social Distancing During the COVID-
19 Pandemic: Who Are the Present and Future Non-compliers? Public Administration
Review, 80(4), 1-32.
Phillips, Emma. (2015). The Changing Dimensions of Privacy in the Workplace: Legal
Rights and Labour Realities. Canadian Labour and Employment Law Journal, 18(2),
467-491.
Rainie, Lee, & Duggan, Maeve. (2016). Privacy and Information Sharing. Washington,
DC: Pew Research Center.
Ravid, Daniel M., Tomczak, David L., White, Jerod C., & Behrend, Tara S. (2020). EPM
20/20: A Review, Framework, and Research Agenda for Electronic Performance
Monitoring. Journal of Management, 46(1), 467-491.
Roberts, Alasdair. (2020). The Third and Fatal Shock: How Pandemic Killed the
Millennial Paradigm. Public Administration Review, 80(4), 1-7.
Schuster, Christian, Weitzman, Lauren, Mikkelsen, Kim Sass, Meyer-Sahling, Jan,
Bersch, Katherine, Fukuyama, Francis, Paskov, Patricia, Rogger, Daniel, Mistree,
Dinsha, Kay, Kerenssa. (2020). Responding to COVID-19 Through Surveys of Public
Servants. Public Administration Review, 80(4), 1-20.
Scott-Hayward, Christine S., Fradella, Henry F., & Fischer, Ryan G. (2015). Does
Privacy Require Secrecy? Societal Expectations of Privacy in the Digital Age. American
Journal of Criminal Law, 43(1), 19-60.
Slobogin, Christopher, & Schumacher, Joseph E. (1993). Reasonable Expectations of
Privacy and Autonomy in Fourth Amendment Cases: An Empirical Look at
“Understandings Recognized and Permitted by Society”. Duke Law Journal, 42(4), 727-
775.
This article is protected by copyright. All rights reserved.
Accepted Article
Smith, Alisa, Madden, Sean, & Barton, Robert P. (2016). An Empirical Examination of
Societal Expectations of Privacy in the Digital Age of GPS, Cell Phone Towers, &
Drones. Albany Law Journal of Science & Technology, 26(1), 111-142.
Steele, Chandra. (2020, February 14th). The Quantified Employee: How Companies Use
Tech to Track Workers. New York, NY: PC Magazine.
Taylor, Frederick W. (1912, 2012). Scientific Management. In Jay M. Shafritz & Albert
C. Hyde (Eds.), Classics of Public Administration (Vol. 7th, pp. 37-43). Boston, MA:
Wadsworth.
Thompson, Clive. (2020, June 9th). What If Working from Home Goes on … Forever?
New York: New York Times Magazine.
Tummers, Lars L., Bekkers, Victor, Vink, Evelien, & Musheno, Michael. (2015). Coping
during Public Service Delivery: A Conceptualization and Systematic Review of the
Literature. Journal of Public Administration Research and Theory, 25(4), 1099-1126.
Weckert, John. (2002). Trust, Corruption, and Surveillance in the Electronic Workplace.
In Klaus Brunnstein & Jacques Berleur (Eds.), Human Choice and Computers: Issues of
Choice and Quality of Life in the Information Society (pp. 109-120). Boston, MA:
Springer.
West, Jonathan P., & Bowman, James S. (2016). Electronic Surveillance at Work: An
Ethical Analysis. Administration & Society, 48(5), 628-651.
Wilson, Melanie D. (2008). The Return of Reasonableness: Saving the Fourth
Amendment from the Supreme Court. Case Western Reserve Law Review, 59(1), 1-60.
Yin, Hong, Biao, Lee, John Chin-Kin, Jin, Y.-L., & Zhang, Z.-H. (2013). The Effect of
Trust on Teacher Empowerment: The Mediation of Teacher Efficacy. Educational
Studies, 39 (1), 13-28.
Young, Kimberly. (2010). Policies and Procedures to Manage Employee Internet Abuse.
Computers in Human Behavior, 26(6): 1467-1471.
This article is protected by copyright. All rights reserved.
Accepted Article
Table 1: Workplace surveillance technologies and their descriptions as presented to respondents
in the surveys (presented in random order).
Key logger: Records how many keys on your computer keyboard were touched per hour
AI email software: Algorithm uses keywords drawn from employee emails to report to bosses about the
office's mood
Keycard with Radio Frequency Identification (RFID): Tracking the location of employees and times in
which workers are in the building
Internet usage: Reports on the websites workers spend time and for how long
OccupyEye: Box under desk that senses body in office to track attendance and body heat measures
Handwashing badge: Sticker worn by employees to track handwashing practices at work
Random photo capture: Computer camera takes photo randomly six times every hour to ensure those working
remotely are at their computer
Clickstream software: Tracks how computer users click and navigate the computer and internet during work
hours
Wellness apps: Incentivized wellness programs at work using FitBit or similar technology to monitor physical
activity.
Non-visible camera: Hidden cameras in workplace to measure the timing of breaks and movements around the
office
Humanyze badges: Analyzes speech through volume and pitch, notes who they spend time with, and maps the
paths of their days
Facial recognition: Monitoring employee activity and enhancing security of the workplace
This article is protected by copyright. All rights reserved.
Accepted Article
Table 2: Workplace surveillance technologies and perceived intrusiveness and reasonableness, ranked.
Technologies
Intrusiveness
(0 – not at all intrusive to 100 – extremely intrusive)
Reasonableness
(1- very unreasonable to 6- very reasonable)
Citizens
web survey
Citizens
phone survey
Public
servants web
panel
Citizens
web survey
Citizens
phone survey
Public
servants web
panel
Average
intrusiveness
Average
intrusiveness
Average
intrusiveness
Median
(and mean)
reasonableness
Median
(and mean)
reasonableness
Median
(and mean)
reasonableness
Non-visible camera
75.6%
---
74.6%
(n.s.)
89.9%
(+14.3%***)
Unreasonable
(m=2.31, ---)
Very
unreasonable
(m=2.36*)
Very
unreasonable
(m=1.59***)
Humanyze badges
74.7%
---
75.4%
(n.s.)
89.6%
(+14.9%***)
Unreasonable
(m=2.38, ---)
Unreasonable
(m=2.35)
Very
unreasonable
(m=1.60***)
Random photo capture
73.1%
---
72.8%
(n.s.)
88.2%
(+15.2%***)
Unreasonable
(m=2.39, ---)
Unreasonable
(m=2.54, n.s.)
Very
unreasonable
(m=1.67***)
OccupyEye
67.3%
---
70.3%
(+3.0%*)
84.4%
(+17.1%***)
Unreasonable
(m=2.57, ---)
Unreasonable
(m=2.56**)
Very
unreasonable
(m=1.71***)
AI email software
68.1%
---
68.7%
(n.s.)
82.9%
(+14.8%***)
Somewhat
unreasonable
(m=2.74, ---)
Unreasonable
(m=2.60***)
Very
unreasonable
(m=1.93***)
Key logger
59.5%
---
65.5%
(+5.9%***)
73.8%
(+14.3%***)
Somewhat
unreasonable
(m=2.96, ---)
Unreasonable
(m=2.76***)
Unreasonable
(m=2.21***)
Clickstream software
58.7%
---
60.7%
(n.s.)
69.4%
(+10.7%***)
Somewhat
unreasonable
(m=3.16, ---)
Somewhat
unreasonable
(m=3.05*)
Unreasonable
(m=2.70***)
Facial recognition
57.6%
---
61.9%
(+4.3%**)
68.9%
(+11.3%***)
Somewhat
unreasonable
(m=3.30, ---)
Somewhat
unreasonable
(m=3.08***)
Unreasonable
(m=2.65***)
This article is protected by copyright. All rights reserved.
Accepted Article
Keycard with Radio Frequency
Identification (RFID)
55.8%
---
56.7%
(n.s.)
64.9%
(+9.1%***)
Somewhat
unreasonable
(m=3.28, ---)
Somewhat
unreasonable
(m=3.25, n.s.)
Unreasonable
(m=2.89***)
Internet usage
55.8%
---
56.1%
(n.s.)
64.4%
(+8.6%***)
Somewhat
reasonable
(m=3.41, ---)
Somewhat
unreasonable
(m=3.28*)
Somewhat
unreasonable
(m=3.85***)
Wellness apps
48.9%
---
63.4%
(+14.5%***)
56.1%
(+7.2%***)
Somewhat
reasonable
(m=3.55, ---)
Somewhat
unreasonable
(m=3.02***)
Somewhat
unreasonable
(m=3.29*)
Handwashing badge
46.9%
---
55.2%
(+8.2%***)
63.6%
(+16.7%***)
Somewhat
reasonable
(m=3.66, ---)
Somewhat
unreasonable
(m=3.44**)
Unreasonable
(m=2.81***)
(n=2,001)
representative
sampling; young
oversampled
(n=1,008)
weighted
representative
sampling
(n=346)
(n=2,001)
representative
sampling; young
oversampled
n=1,008)
weighted
representative
sampling
(n=346)
(***, p>0.001; **, p>0.01; *, p>0.05)
This article is protected by copyright. All rights reserved.
Accepted Article
Figure 1. Trust in colleagues, trust in others and survey mode effects on aggregated intrusiveness and reasonableness scores
(citizen web panel as the base of comparison), by experimental condition (CI = 95%).
This article is protected by copyright. All rights reserved.
Accepted Article
Figure 2. Differences between the intrusiveness and reasonableness scores of twelve work surveillance technologies according
to young (18-30year old; n=1,169); and older (30+year; n=1,840) respondents (representative web panel and weighted-
representative phone survey data pooled).
This article is protected by copyright. All rights reserved.
Accepted Article
This article is protected by copyright. All rights reserved.
Accepted Article
Table 3: Typology of the surveillance-performance link and views of unreasonable work surveillance technologies.
Median respondent on reasonableness (public servants)
Clear surveillance
performance link
Somewhat unreasonable or
unreasonable
Very unreasonable
Yes
Key logger
Clickstream software
Internet usage
-
No
Facial recognition
Keycard w/ RFID
Handwashing badge
Wellness apps
Non-visible camera
Humanyze badges
Random photo capture
OccupEye
AI email software
This article is protected by copyright. All rights reserved.
Accepted Article
Appendix A
Citizens
web survey
Citizens
phone survey
Public servants
web panel
(n=2,001)
representative
sampling; young
oversampled
(n=1,008)
weighted
representative
sampling
(n=346)
Age
18-30
50.2% (1005)
17.6% (177)
13.3% (46)
31-40
14.3% (284)
22.6% (228)
29.2% (101)
41-54
14.4% (297)
24.0% (242)
35.6% (123)
55+
20.7% (415)
35.7% (361)
22.0% (76)
Gender
Female
50.7% (1015)
44.4% (447)
53.8% (186)
Male
48.4% (968)
49.1% (495)
44.2% (153)
Non-binary,
transgender, rather not
say
1.4% (18)
6.6% (66)
2.0% (7)
Canadians
come
from all
over the
world.
What is
your
ethnic
origin?
European
59.5% (1191)
62.6% (631)
68.5% (235)
Black
3.8% (75)
4.5% (45)
1.5% (5)
South Asian
6.5% (129)
3.5% (35)
4.1% (14)
East Asian
7.4% (148)
1.6% (16)
5.3% (18)
Latin America
2.0 % (40)
1.3% (43)
2.0% (7)
Indigenous
2.0 % (40)
3.6% (37)
1.2% (4)
other
11.9% (239)
7.2% (72)
10.8% (37)
Prefer not to say
7.0% (139)
15.7% (158)
6.7% (23)
Education
High school / GED
34.5% (691)
17.3% (175)
0.9% (3)
College diploma or
vocational training
28.8% (576)
30.1% (304)
7.2% (25)
Undergraduate degree
25.8% (517)
29.6% (299)
29.2% (101)
Masters or add.
Professional training
9.0% (180)
18.6% (187)
56.1% (194)
Medical or doctorate
degree
1.9% (37)
4.3% (44)
6.7% (23)
This article is protected by copyright. All rights reserved.
Accepted Article
