Abstract and Figures

Ban-the-Box (BTB) legislation, which bans employers from asking about criminal history records on the initial job application, is arguably the most prominent policy arising from the prisoner reentry movement. BTB policies assume: 1) most employers ask about criminal records, and 2) inquiries occur at the application stage. However, we lack reliable information about the validity of these assumptions or about public attitudes towards background checks, which limits our understanding of the potential scope of this innovative policy. Using survey data from a national probability sample, we estimate that in the past year, over 31 million U.S. adults were asked about a criminal record on a job application. According to our survey, virtually all of the criminal record inquiries occurred at the application stage, highlighting the potential of BTB. However, we also found that the public is sharply divided on whether to prevent employers from asking on applications, as per BTB.
Content may be subject to copyright.
Criminal Records and Employment: A Survey of
Experiences and Attitudes in the United States
Forthcoming in Justice Quarterly
Megan Denver*
Justin T. Pickett
School of Criminal Justice
University at Albany, SUNY
Shawn D. Bushway
Rockefeller College of Public Affairs & Policy
University at Albany, SUNY
Ban-the-Box (BTB) legislation, which bans employers from asking about criminal
history records on the initial job application, is arguably the most prominent policy arising from
the prisoner reentry movement. BTB policies assume: 1) most employers ask about criminal
records, and 2) inquiries occur at the application stage. However, we lack reliable information
about the validity of these assumptions or about public attitudes towards background checks,
which limits our understanding of the potential scope of this innovative policy. Using survey data
from a national probability sample, we estimate that in the past year, over 31 million U.S. adults
were asked about a criminal record on a job application. According to our survey, virtually all of
the criminal record inquiries occurred at the application stage, highlighting the potential of BTB.
However, we also found that the public is sharply divided on whether to prevent employers from
asking on applications, as per BTB.
Keywords: Ban-the-Box, hiring, employment, conviction histories, public opinion
* Direct correspondence to Megan Denver, School of Criminal Justice, University at Albany, SUNY, 135 Western
Avenue, Albany, NY 12222; e-mail: mdenver@albany.edu. This research was funded by a grant to the Center for
Social and Demographic Analysis (CSDA) from the National Institute of Child Health and Human Development
Policies focused on prisoner reentry are an excellent example of the dynamic nature of
criminal justice policy in the United States. Prisoner reentry did not enter the criminal justice
conversation as a separate platform worthy of policy focus until 1999 when Jeremy Travis, first
as the director of the National Institute of Justice, and then as a senior fellow at the Urban
Institute, reinvented the term and began a concerted effort to draw attention to the large numbers
of prisoners returning to U.S. communities every year (Toney, 2007; Travis, 2005).1 A year later,
then U.S. Attorney General Janet Reno (2000, p. 1) pointed to prisoner reentry as “one of the
most present problems we face as a nation.” The effort has had two primary points of emphasis –
creating programs and services for those reentering the community after prison, and reducing
barriers faced by individuals with criminal records (Petersilia, 2003; Travis, 2005). By any
reasonable measure, the effort has been quite successful, with new national advocacy
organizations like the National H.I.R.E. Network, and government reentry councils at the federal,
state and local level dedicated to finding ways to improve the reentry process and reduce
recidivism (Federal Interagency Reentry Council, 2016).
Not surprisingly, research efforts on new hot topics take a while to catch up, and policy
innovations often proceed without the benefit of research. For example, Justice Quarterly has
published a large number of papers—55—on reentry in its history, but most of them (39) have
been published since 2009.2 Most of the attention has focused on evaluating programs designed
to improve prisoner reentry, perhaps as a result of federal funding set aside for evaluating large-
scale initiatives like the Serious and Violent Offender Reentry Initiative (SVORI) and the Second
Chance Act. Correspondingly, researchers have begun to explore public attitudes toward these
1 John Irwin discussed reentry as a separate stage of the prisoner experience in 1970 (Irwin, 1970), a fact that Travis
learned only after he started using the word (Toney, 2007).
2 Only six papers were published that fall under that topic prior to 2000, and none of these papers used the word
reentry to describe their research (the papers were archived online in 2006, when reentry was used as a keyword).
reentry policies (Garland, Wodahl, & Saxon, 2017; Garland, Wodahl, & Schuhmann, 2013;
Ouellette, Applegate, & Vuk, 2016).
Research on the benefits of reducing barriers to reentry, often known as collateral
consequences, is less common, although researchers and legal scholars have documented the
massive influx of laws and regulations that create barriers to employment, education, housing,
civic rights, and government assistance (The Council of State Governments, 2017). Such
restrictions may generate a cycle of stigma, as individuals with criminal records collect feedback
—both explicit and implicit—from employers or other decision makers that serve as gatekeepers
to opportunities (Pager, 2007).3
Despite the clear evidence that these barriers exist (Pager, 2003), there are only a few
(very recent) evaluations of the oft-repeated claim that employment barriers, such as criminal
background check denial decisions, lead directly to increased recidivism (Author Citation A;
Siwach, 2016). Moreover, there is little research demonstrating that policies designed to increase
access to employment by altering when and how criminal records are introduced and used in the
hiring process reduce recidivism.4 This lack of research is particularly problematic if there are
potential unintended consequences created by such policies.
The possibility that unintended consequences may result from new policy initiatives has
recently become very real for one of the more successful policy movements aimed at reducing
collateral consequences, known as “Ban-the-Box” (BTB) (Mullainathan, 2016; Semuels, 2016).
BTB seeks to afford greater employment opportunities for individuals with conviction histories
3 A symbolic interaction perspective (Blumer, 1969; Mead, 1934) posits that individuals are influenced and shaped
by others’ perceptions, negative reactions, and formal labels to violations of prosocial norms (Lemert, 1951;
Tannenbaum, 1938). Therefore, when an individual internalizes perceptions of failing, the negative feedback can
induce a self-fulfilling prophecy (Goffman, 1963) and lead to recidivism.
4 For an exception, see Author Citation C, which evaluates a criminal background check decision guideline that
increases clearance to work for individuals with “older” conviction records. In addition, Author Citation B and
Siwach (2016) show clear heterogeneous treatment effects. Policies designed to decrease recidivism must operate
on those people for whom employment will reduce recidivism.
by delaying criminal record inquiries until later in the hiring process, after applicants’
qualifications are first investigated. Over the past decade, the federal government, nearly half of
all U.S. states, and over 100 cities and counties have adopted BTB legislation5, which blocks
employers from asking about job applicants’ criminal records on employment applications
(Rodriguez & Avery, 2016). However, recently, a number of studies have documented evidence
of increased discrimination against Black applicants when employers lack early access to
criminal records as a result of BTB rules, creating a difficult tradeoff for policy makers between
helping those with criminal records and potentially creating more racial disparity in employment
outcomes (Agan & Starr, 2016; Doleac & Hansen, 2016; Vuolo, Lageson, & Uggen, 2017).
It is possible that such discrimination is partially motivated by ambivalence on the part
of the public and decision makers about efforts to reintegrate individuals with criminal records
back into the labor market. In fact, ambiguity often sparks a policy decision in the first place.
Having multiple available options when considering the same issue “facilitates taking
appropriate action and shaping preferences without a priori estimating the consequences”
(Zahariadis, 2014, p. 25). In his Multiple Streams Approach, Kingdon (1995) suggests when
three key factors converge—problems, potential policy solutions, and politics—there is a
heightened opportunity for policy change. To date, however, the BTB movement and associated
policy changes have proceeded in the dark with respect to an important component of Kingdon’s
(1995) third key factor: relevant public experiences and attitudes. And, according to Garland,
Wodahl, and Cota (2016, p. 1406), “a lack of public support for prisoner reentry initiatives could
undermine the sustainability of prisoner reentry as a large-scale movement.”
5 BTB policies initially focused on public employers (Rodriguez & Avery, 2016), although the policies have evolved
in scope over time and extended to the private sector in some states and cities (Agan & Starr, 2016). In
addition, several large private employers—including Target, Walmart, Home Depot, Bed Bath and Beyond and Koch
Industries—have adopted BTB nationally (Levine, 2015).
The first step to better understanding reentry policy strategies that focus on employment
opportunities is to start with documenting the scope of the criminal background check issue that
BTB seeks to affect. Although several researchers and companies have made localized attempts
to quantify the extent of criminal record inquiries in the hiring process, there are not currently
any nationally representative estimates. National estimates should ground efforts to reduce the
potential harm for this practice, and create a better understanding of the potential target audience
for policies such as BTB.
There are also no national estimates of the degree to which the criminal background
check process is initiated at the application stage, the specific target of Ban the Box policies.
Only a few case studies have been able to estimate whether and when job applicants are either
asked about their criminal records or have criminal background checks conducted. These are two
different processes that are often described interchangeably. BTB policies are focused on
employers’ criminal record inquiries for job applicants, more so than external criminal
background checks. In the present study, we explore the timing of criminal record inquiries at the
national level and whether this has shifted over time, which has direct implications for the reach
and potential effects of policies targeting specific points in the process of applying for a job.
A final issue is that there is very limited information about whether the public supports
criminal record inquiries in the hiring process, and, if so, at what stage of the process they
believe these inquiries should first occur. Thus, it remains unclear whether U.S. residents are
receptive to the types of policy reforms advocated under the BTB framework. Not least, we
currently lack knowledge about whether attitudes toward criminal record inquiries in hiring vary
across different subgroups in the population. There are large racial and gender differences in
criminal justice experiences (Travis, Western, & Redburn, 2014), and thus it would not be
surprising if there were similar gaps in attitudes about criminal record checks.
In the present study, we examine the public’s experiences with criminal record inquiries
in hiring, documenting both the prevalence and timing of these inquiries. We then explore the
extent and correlates of public support for criminal record inquiries in hiring. Below, before
describing our methods and findings, we first review the extant evidence on the prevalence and
consequences of criminal records, discuss the “Ban-the-Box” movement, and summarize the
relevant public opinion literature.
Prevalence and Consequences of Criminal Records
Criminal records are pervasive in the United States, due in large part to policy choices
that resulted in the “mass incarceration” of people (Alexander, 2012; Pager, 2007). Over the past
four decades, the U.S. incarceration rate has quintupled (Travis et al., 2014), and in 2014,
approximately 1 in 36 adults were under some type of criminal justice supervision (Kaeble,
Glaze, Tsoutis, & Minton, 2015). Extant evidence shows that almost a third of individuals in the
United States are arrested by age 23 (Brame, Turner, Paternoster, & Bushway, 2012). Racial
minorities and men are notably overrepresented among this large population of criminal record-
holders (Western, 2006); for example, around 49% of black males have been arrested by age 23,
compared to 38% of white males (Brame, Bushway, Paternoster, & Turner, 2014).
The widespread involvement of U.S. adults in the criminal justice system has important
implications for the collateral consequences this group faces. In the employment context, job
applicants with known criminal records are seen as less desirable job candidates to employers
than other stigmatized groups (Holzer, Raphael, & Stoll, 2002). In a series of seminal studies,
Pager (2003, 2007) found that applicants with criminal history records experience fewer job
callbacks than otherwise identical individuals without records, especially if the applicants are
African American. More recently, Uggen, Vuolo, Lageson, Ruhland, and Whitham (2014)
conducted an audit study in Minneapolis-St. Paul that was modeled after Pager’s (2003) study
design, but the research team substituted a misdemeanor arrest in place of the felony conviction
record. Uggen et al. (2014) find smaller effect sizes, although employer callbacks are still
statistically significantly lower for criminal record holders.
The majority of employers report that they would not knowingly hire applicants with
criminal records (Pager, 2007, p. 34). Many employers doubt criminal record holders’ general
trustworthiness (Pager & Quillian, 2005). They also worry about legal liability if the job
candidate goes on to commit a crime (Lageson, Vuolo, & Uggen, 2015), and want to ensure a
safe work environment (Society for Human Resource Management, 2012). These concerns are
typically exacerbated if the job position involves interacting with vulnerable people or sensitive
information (Watstein, 2009), although the ability to easily conduct criminal background checks
has also expanded their use to “even menial” jobs (Solove & Hoofnagle, 2006).
At the same time, employment is important for establishing independence and
transitioning to adulthood (Furstenberg, Kennedy, McLoyd, Rumbaut, & Settersten, 2004), and is
theoretically linked to reduced recidivism for individuals with criminal records (e.g., Laub &
Sampson, 2001; Uggen, 2000). For those with criminal records who are motivated to work, a
criminal background check denial due to their record can lead to reduced employment for several
years (Author Citation B) and increased levels of recidivism (Author Citation A), relative to
comparable individuals with criminal records who pass the check.
The “Ban-the-Box” Movement
Ban-the-Box (BTB) is one proposed policy solution to help individuals with criminal
records—and minority job applicants in particular—gain employment. BTB does not prevent
employers from asking about criminal records—rather, it forces the employer to wait until after
the application stage to ask. The goals of this delay are two-fold. First, job applicants with
records may be more likely to apply to jobs that withhold the question from the application.
Second, employers might be more willing to consider the applicant for employment if they only
find out about the applicant’s record after having already reached some positive conclusions
about the applicant’s qualifications (Vuolo et al., 2017). Therefore, delaying criminal record
inquiries may increase the likelihood that individuals with criminal records, who are
disproportionately racial and ethnic minorities, are at least able to get their foot in the door when
applying for jobs.
Hawaii first adopted BTB in 1998, but the policy did not gain immediate traction.
Advocates proclaimed BTB a “movement” in 2004 (The Southern Coalition for Social Justice,
2014), and other states started joining BTB by 2009 (Rodriguez & Avery, 2016). BTB is now
rapidly being adopted; President Obama directed the federal Office of Personnel Management to
implement BTB in 2015 (White House, 2015), and 10 states enacted BTB just between January
2015–June 2016 (Rodriguez & Avery, 2016). As a result of these legislative initiatives, over half
of the U.S. population currently lives in a BTB jurisdiction (Rodriguez & Avery, 2016).
Several recent studies have evaluated the effects of BTB. As intended by the policy,
callbacks for interviews generally increase for individuals with criminal records after BTB takes
effect (Agan & Starr, 2016). The percentage of new job hires with criminal records increased
after BTB was passed in Durham, North Carolina (The Southern Coalition for Social Justice,
2014). Additionally, BTB might improve employment for low-skilled African Americans, who
are more likely to acquire criminal records, and residents in high-crime neighborhoods (Shoag &
Veuger, 2016). However, researchers have also documented several unintended consequences. In
line with statistical discriminatory theory (Bushway, 2004), the black-white gap in interview
callbacks widens when employers are denied upfront information about criminal history records
(Agan & Starr, 2016; Vuolo et al., 2017) and young black and Hispanic men without college
degrees experience reduced employment (Doleac & Hansen, 2016). In other words, in lieu of
criminal record information employers may rely on stereotypes and observable information to
estimate job applicants’ work productivity and/or probability of committing new crimes. This
causes employment barriers to shift and leads to a difficult tradeoff between broadly increased
hiring for individuals with criminal records and increased disadvantages for certain groups.
While BTB policies specifically target the application stage of the hiring process, as
discussed below, extant evidence about the prevalence and timing of criminal record inquiries is
limited. Therefore, the potential scope of this type of policy is currently unknown. Second, as
policymakers grapple with the potentially complicated effects of BTB, public opinion research
on the use of criminal records in hiring—especially given the number of people possibly
impacted by BTB policies—would provide valuable information for the continued debate.
Prior Research
on the Prevalence of Criminal Record Inquires and Checks
There is very little available evidence on the extent to which employers actually ask
about criminal records when hiring, or at which point in the process they ask. The current
evidence includes a handful of case studies from different locations and time periods. One of the
most oft-cited estimates of criminal recording checking is from a professional society’s survey of
approximately 400 member organizations. The survey indicates that in 2012, almost 70% of
employers conducted criminal background checks at some point during the hiring process on all
job candidates; the figure rises to 87% when including organizations that conduct checks for at
least some job positions (Society for Human Resource Management, 2012). Of the 343 member
organizations conducting some type of criminal background check in 2012, 62% report
conducting the check “after a contingent job offer,” and a third reported doing so before the job
offer but after an interview (Society for Human Resource Management, 2012). However, the
Human Resource survey does not ask respondents whether and when they first ask about
criminal records. Since criminal record information is used to sort applicants before a formal
criminal background check is conducted (Pager, 2003; Vuolo et al., 2017), a useful analysis
would be to determine whether and when people are first asked.
An employer survey in Los Angeles in 2001 found similar, albeit slightly lower,
estimates. Out of business establishments that recently hired at least one non-college degree
applicant, 63% reported “checking” criminal records, although it was not clear when during the
hiring process the inquiry or formal background check occurred (Holzer, Raphael & Stoll, 2006).
An advantage to both the Human Resource member survey and Los Angeles employer survey is
that they span across industry type and firm size, both of which are meaningful correlates of
conducting criminal background checks (Holzer, Raphael & Stoll, 2006). However, it is unclear
what types of employers select into the professional society’s membership and opt to take the
survey. In addition to focusing solely on Los Angeles, the employer survey estimates are also
potentially outdated. 6 Most notably, the survey occurred during 2001, and does not capture the
6 Although the Multi-city Study of Urban Inequality data that Holzer, Raphael and Stoll (2006) use also include
Atlanta, Boston, and Detroit, Los Angeles is the only city with a second survey in 2001; the other city surveys were
conducted in the early 1990s.
rapid increase in criminal background checks in the U.S. after the September 11th terrorist attacks
(SEARCH, 2005).
More recently, Vuolo et al. (2017) collected job applications as part of their 2007-2008
audit study in Minneapolis to examine whether, and how, employers ask about criminal records.
Targeting entry level job positions, the study found 78% of employers asked about criminal
records on the application (Vuolo et al., 2017). As the authors note, there is an important
distinction between asking applicants about criminal records and conducting criminal
background checks, particularly when considering application screening processes (e.g., Pager,
2003). There are not clear links in the general employment literature between asking about
criminal records and conducting criminal background checks, and only limited evidence in the
licensing literature. For example, through interviews with state licensing boards for barbers,
Ewald (2016) finds that not all licensure officials actually conduct formal criminal background
checks, fully relying on applicant self-disclosure instead. In other words, rather than both asking
applicants about their records and then following up with an external check, employers may stop
after the first step. BTB assumes this “first step” check typically happens during the application
stage, leading to the premature dismissal of potentially qualified applicants.
There are two major challenges in making inferences from prior studies to the broader
population of employers or the public. First, “checking a criminal record” may mean different
things to different employers. The term may be used interchangeably to describe asking about a
criminal record on a job application and conducting a more formal (and more expensive) external
check. The distinction is important for understanding the scope of the issue, its relevance to
BTB, and the potential implications for job applicants. Second, it can be difficult to define a
population of employers, particularly since smaller companies or organizations may not be static
over time or easy to identify, and there is not an “average” job that would speak to most job
applicants’ experiences. Prior research has therefore been localized to a particular set of
employers or a particular geographical location. An alternative approach, which we use here, is
to generate an estimate of the prevalence and timing of criminal record inquiries from a
representative sample of U.S. residents, based on their most recent job search. While not without
limits (e.g., recall errors), this method broadens the scope of the analysis beyond any specific job
type, position level, or geographic location.
Prior Research on Public Attitudes
Unfortunately, despite the importance of prisoner reentry, only a few studies have
explored public support for reentry initiatives using representative samples, and most often they
have relied on data from one of two statewide polls. A series of studies by Garland and
colleagues (2013, 2015, 2016, 2017) have analyzed data from a 2008 survey of 386 residents of
Missouri. Although not focusing specifically on criminal background inquiries or checks, these
studies revealed that Missourians are highly supportive of prison reentry initiatives. At the same
time, most Missourians believed that “during hard economic times, employers should give
preference in their hiring decisions to people who have no prison record over those who have
been in prison” (Garland et al., 2013, p. 37). Garland and colleagues (2015, 2016) also found that
whereas demographic factors were weakly and inconsistently related to views about reentry,
political conservatism was consistently negatively related to support for reentry efforts.
Ouellette et al. (2016, p. 16) analyzed data from a 2014 survey of 165 South Carolinians.
They found that support was “widespread, but … not robust,” particularly when programs were
associated with higher taxes. The majority of South Carolinians supported employment
assistance programs for released prisoners, and indicated that they would not mind working with
people released from prison. However, South Carolinians were split on whether people without
records should be given hiring preference during hard economic times. Like Garland et al. (2015,
2016), Ouellette et al. (2016) found that respondent demographics were inconsistently related to
attitudes toward reentry. However, Ouellette et al. (2016) did not include a measure of political
ideology. Unfortunately, as with the Missouri survey, the poll of South Carolinians did not
directly ask about criminal background inquiries in hiring.
To our knowledge, only one previous study has assessed public attitudes toward criminal
background checks in hiring, but it is dated. The study was conducted by SEARCH and the
Bureau of Justice Statistics (BJS) in 2000 and focused on whether employers should be able to
check (through external sources) applicants’ records (Bureau of Justice Statistics, 2001). In a
telephone survey using a national probability sample (N =1,030 U.S. adults), the researchers
found that over 90% of respondents believed that either all employers (40%) or some employers
depending on the job (55%) should be allowed to request conviction records from government
agencies (Bureau of Justice Statistics, 2001). The study also reported bivariate evidence that
White persons, older persons, the better educated, and those with higher incomes tended to be
more supportive of criminal background checks. However, the study provided no information
about public preferences regarding the timing of record checks, or about whether the public
supports allowing employers to ask the applicants themselves about their criminal records. These
research questions are the most relevant to current BTB policy, which focuses on the application
The paucity of research on the extent and correlates of public attitudes towards criminal
record inquires in hiring is surprising, given the increased attention towards the issue and
proliferation of BTB policies in recent years (Jacobs, 2015; SEARCH, 2005). Public opinion can
gauge the political climate and motivate relevant policy agendas (Thielo, Cullen, Cohen, &
Chouhy, 2015), and policymakers often consider and debate reforms and expenditures with
attention to public attitudes (Nagin, Piquero, Scott & Steinberg, 2006). Indeed, Canes-Wrone and
Shotts (2004) report evidence that politicans are more responsive to public opinion on criminal
justice than other issues. More generally, studies consistently find that popular attitudes help
explain shifts in criminal justice policy over time (Enns, 2016; Nicholson-Crotty, Peterson &
Ramirez, 2009).
For these reasons, we provide the first nationally representative investigation of the
extent and correlates of support for criminal record inquiries in hiring since the 2000 BJS study.
Based on previous statewide findings for public attitudes toward reentry iniatives (Garland et al.,
2013; Ouellette et al., 2016), we expect the public to be supportive of the BTB effort to ban
criminal record inquries from job applications. There is insufficient evidence to formulate
hypotheses about the demographic correlates of such support. However, because there are large
racial and gender differences in criminal justice experiences (Travis et al., 2014), it would not be
surprising to observe similar differences in attitudes. One realtively consistent finding from the
extant literature on public attitudes is that right-wing political ideology reduces support for
reentry initiatives (Garland et al., 2015, 2016). Therefore, we expect Republicans to be less likely
than their political counterparts to support BTB strategies, and more likely to support
application-stage criminal record inquires.
We included a battery of questions on criminal record inquiries during the hiring process
in an omnibus survey administered by the GfK Group (formally Knowledge Networks) to a
representative sample of 1,023 adult (18 and older) residents of the United States. The sample
size in the analysis ranges from 962 to 1,009 because of item nonresponse. The margin of error
for the survey is +/- 3 percentage points. All of the respondents were members of GfK’s
probability-based “KnowledgePanel Omnibus.” Recruitment into the panel occurred through
random sampling of telephone numbers and residential addresses. The coverage rate for this
sampling procedure is 97% of American households. The omnibus survey was conducted online.
Panelists who lacked a computer or Internet access were provided with both. GfK Knowledge
Networks is one of the leading and most highly trusted polling platforms for conducting
probability-based Internet surveys of the general U.S. population (Weinberg, Freese, &
McElhattan, 2014). Allcott (2011: 99) explains that “Knowledge Networks … maintains perhaps
the highest-quality publicly available survey platform.” Chang and Krosnick (2009, p. 641)
document that the probability-based Internet surveys conducted by GfK Knowledge Networks
manifest “the optimal combination of sample composition accuracy and self-report accuracy.”
The GfK Group fielded the omnibus survey during a three-day period in June 2016,
during which they sent one follow-up email to the invited panelists to encourage participation.
The American Association for Public Opinion Research’s (AAPOR) Standard Definitions (2016,
pp. 48-49) provides formulas for calculating the final outcome rates in surveys with pre-
recruited, probability-based online panels. Per these calculations, the survey completion rate was
29%, indicating that more than 1 in 4 of the panelists who were invited to participate in this
particular omnibus survey did so. The initial recruitment rate into the panel was 12.8%, meaning
that roughly 1 in 8 of those individuals who GfK contacted (on an earlier date) and asked to join
the panel did so. The profile rate was 64.6%, meaning that upon recruitment, most of those who
agreed to join the panel completed all of the necessary baseline information about their
household to enroll in the panel. The AAPOR defines the cumulative response rate as a
multiplicative function of these three rates (recruitment rate x profile rate x completion rate =
cumulative response rate). In our survey, this formula yields a cumulative response rate of 2.4%
(.128 x .646 x .29 = .024). This cumulative response rate is very similar to those reported in other
recent studies conducted with the KnowledgePanel (e.g., Tourangeau, Maitland, & Yan, 2016).
Despite the use of probability sampling, some readers may question whether our sample
can be representative of the American public given the 2.4% cumulative response rate. Simply
put, nonresponse bias may undermine generalizability. However, in a recent report on survey
research methods to the National Science Foundation, Krosnick et al.’s (2015, p. 6) concluded
that “nonresponse bias is rarely notably related to [the] nonresponse rate.” Their conclusion
echoes meta-analytic findings showing that 1) there is a weak relationship between response
rates and nonresponse bias, and 2) most of the variation in nonresponse bias exits within studies
(across different estimates), rather than between studies (as a function of study characteristics
like the response rate) (Groves, 2006; Groves & Peytcheva, 2008). These findings reflect the fact
that nonresponse bias occurs only when the propensity to respond is correlated with the specific
survey variables, and not simply when the response rate is low.
Notably, a correlation between response propensity (P) and survey variables (Y) emerges
either when 1) the latter affects the former (Y → P, e.g., interest in topic drives participation), or
2) both share a common cause (P ← Z → Y, e.g., demographics influence participation and
attitudes) (Groves et al., 2009). In the first case (Y → P), nonresponse is “nonignorable.”
However, in the second case (P ← Z → Y), after controlling for (e.g., weighting by) the common
cause (Z), nonresponse is “missing at random” (Groves et al., 2009). We do not believe
nonresponse bias is a major concern in our study because it is unlikely that nonresponse in our
survey is nonignorable (Y → P), given that 1) most of the nonresponse occurred before our
actual survey (i.e., at the recruitment and profile stage), and 2) our questions were embedded in
an omnibus survey where the survey invitation excluded any mention of criminal background
checks and most of the other questions were unrelated. In addition, the data are weighted to
adjust for population benchmarks from the U.S. Census and Current Population Survey, which
should render nonresponse “missing at random” to the extent that it is a function of the
respective demographic characteristics. Population parameters along with the unweighted and
weighted demographic characteristics for all respondents in the sample with complete data on the
respective demographic items are provided in Supplement A.
It also bears nothing that Groves and Peytcheva’s (2008, p. 177) meta-analysis showed
that nonresponse bias tended to be higher in interviewer-administered surveys, government-
sponsored surveys, and surveys where respondents lacked prior involvement with the sponsor.
By contrast, our survey was self-administered, lacked government sponsorship, and was
completed by panelists who all had prior involvement with the surveyor (GfK).7 Moreover, a
large portion of our analysis focuses on multivariate relationships. Although nonresponse bias
often has sizeable effects on univariate statistics (especially in the absence of weighting), it most
commonly has relatively small effects on relationships between variables (correlation deviations
of .05 or less) (Heggestad et al., 2015). For example, in a rigorous analysis of nonresponse bias,
Amaya and Presser (2017, p. 34) found that “unlike the univariate bias, the bias observed in the
multivariate models was often small and rarely changed the interpretation of the model.” Finally,
7 Groves and Peytcheva (2008) also found that nonresponse bias was higher in general population surveys and for
attitudinal (versus behavioral) estimates. However, these design characteristics are necessarily set by the research
objective, and are obligatory in surveys like ours where the goal is to study attitudes in the general population.
as noted below, we are able to replicate one set of the findings from the GfK sample with an
independent sample collected through different means. As Rogelberg and Stanton (2007, p. 202)
argue, “replicating a set of findings across multiple data samples is another compelling method
of demonstrating an absence of substantive nonresponse bias.”
There are three questions of interest for our study. The first item identifies when the
respondent last applied for a job. The survey question was: “How long has it been since you last
applied for a job?” Response options included: “I have never applied for a job,” “less than a
year,” “1-2 years,” “3-4 years,” “5-10 years,” or “more than 10 years.” Only 19 respondents
reported never having applied for a job8.
Responses to the first question provide the context for our second question, which
measures respondents’ experiences with criminal record inquiries during their most recent job
search: “Please think about the last time you applied for a job. When did the employer first ask
you if you had ever been convicted of a crime, or were you never asked?” The four response
options were: “At the initial application stage (i.e., on the application form)”; “at the interview
stage, before the final hiring decision”; “after the final hiring decision, before beginning work”;
or “never, the employer didn’t ask.” An advantage of collecting self-report information on
criminal record inquiries, as opposed to self-reported formal criminal background check
experiences, is that job applicants should be aware of whether, and when, they are asked about a
criminal record as part of the job application. In addition, under the Fair Credit Reporting Act
(FCRA), employers are prohibited from conducting criminal background checks without first
8 These individuals are excluded from models examining respondent experiences with criminal background checks,
but are included in analyses assessing support.
notifying individuals about their intention to do so. FCRA also requires that employers who use
information from the criminal background check as part of their decision not to hire an applicant
provide the information used to make that decision to the applicant.9
The final question measures respondents’ views about when, if at all, employers should
be allowed to ask job applicants about their criminal records.10 The survey question was:
A criminal record is a person’s history of ever having been convicted of a crime,
regardless of the sentence (e.g., prison, probation, etc.). Now please think about the hiring
process. In your view, when should employers FIRST be allowed to ask about a job
applicant’s criminal record, or do you think they should never be allowed to ask?
Similar to the question about personal experience, the response options were: “At the
initial application stage (i.e., on the application form)”; “at the interview stage, before the final
hiring decision”; “after the final hiring decision, before the person begins work”; or “never,
employers shouldn’t be allowed to ask about criminal records.” Responses to this question have
direct relevance to the BTB policies, which are foremost concerned with the timing of criminal
record inquiries in hiring—specifically, banning criminal record inquiries from the application
stage. The order in which respondents received the questions about experience with versus
support for criminal record inquiries was randomly rotated.
9 https://www.consumer.ftc.gov/articles/0157-employment-background-checks
10 In the question, we focused on conviction histories for two reasons. First, recent research demonstrates that
conviction records dominate questions on job applications (Vuolo et al., 2016, Table 1). Second, the Equal
Employment Opportunity Commission (2012) strongly advises employers to consider convictions, and not arrests, in
the hiring process.
How commonly are members of the public asked about their criminal records when
applying for jobs? At what stage of the hiring process do such inquiries most often occur? Table
1 presents data on the prevalence of criminal record inquiries across hiring stages for
respondents’ most recent job search. Five findings stand out. First, a sizable percentage of U.S.
residents (21.6%) report applying for a job in the past year. The U.S. Census’s most recent
estimates place the size of the adult U.S. population at 247,773,709. Per our findings, this would
suggest that well over 53 million U.S. adults applied for a job in the past 12 months.
Second, most respondents—58.4% overall, and 71.1% of those who applied for a job
within the past year—say they were asked whether they had a criminal record at some point
during their last job search. The national data thus corroborate findings from prior case studies of
employers. Third, the majority of U.S. residents report that the criminal record inquiry occurred
at the application stage of the hiring process. Indeed, among the 21.6% of U.S. adults who
applied for a job in the past year, 59.2% reported being asked about their criminal record on the
application form. This suggests that in just the past year, nearly 13% of U.S. adults—more than
31 million U.S. residents—applied for a job and were asked about their criminal record on the
job application. This provides evidence that the potential reach of BTB application-stage bans on
criminal record inquiries is considerable.
Fourth, the disaggregated analysis shows that the above pattern of findings emerges for
respondents who applied for a job within the past ten years, regardless of when during that time
period their job search occurred. Finally, a different pattern of results is observed for members of
the public who applied for a job more than ten years ago. Most of these individuals report not
being asked about their criminal record. This finding is consistent with the argument that the use
of criminal record inquiries in hiring has increased over the past few decades (Vuolo et al.,
2017). However, it could also reflect errors in recall due to the extended time frame. Regardless,
it is clear that criminal record inquiries are a prevalent component of the hiring process, and
almost always happen at the application stage. Therefore, Ban-the-Box policies that restrict
questions on the application have the potential to radically change the landscape of criminal
record inquiries in the United States.
For the subsequent analysis, given the theoretical and policy importance of application-
stage criminal background check inquiries (Vuolo et al., 2017), we distinguish the application
stage from all other categories. Table 2 presents results from a logistic regression model
predicting whether respondents were asked about their criminal records on the application form
in their last job search. The outcome variable is coded “1” if the respondent reported being asked
on the application form, and coded “0” if he or she was asked at a later stage or was not asked at
all. The independent variables are coded as specified in Supplement A. The results show that net
of other factors, U.S. residents who applied for jobs within the past ten years were significantly
more likely than those who applied for jobs in earlier time periods to encounter application-stage
questions about their criminal records. The one exception is for those who applied within the past
year; the coefficient for this group, although positive, did not reach the conventional threshold
for statistical significance. This is consistent with the possibility that BTB policies have begun to
have an effect, reducing the use of application-stage criminal record inquiries.
In addition, we find evidence that females, younger persons, those with higher
educational attainment, and individuals residing in the Midwest or West are significantly more
likely to report being asked about their criminal record at the application-stage in their last job
search. The associations for gender, age, and education likely reflect the influence of these
factors on the types of jobs to which these individuals apply. For example, women
disproportionately dominate jobs in the healthcare, personal care and services, and education
industries (Bureau of Labor Statistics, 2015), which may be especially likely to involve
application-stage criminal record inquiries.11
Tables 1 and 2 about here
We now turn our attention to a related set of questions: Do U.S. residents support
allowing employers to ask job applicants about their criminal records? If so, when do
respondents believe this inquiry should first occur? Figure 1 presents the weighted distribution
of responses to the survey question asking if and when the criminal record inquiry should be
allowed. First, the overwhelming majority (91%) of U.S. residents support allowing employers to
ask applicants about their criminal records. Second, and most important for BTB efforts, the
public is roughly split on whether the criminal record inquiry should first occur at the application
stage, or should occur at a subsequent stage of the hiring process, as required under BTB
policies. A slight majority (57%) of U.S. residents believe job applicants should first be asked
about their criminal records on the application form, but a sizeable minority (34%) believes the
inquiries should be delayed until either the interview stage or after an offer is made.
Figure 1 about here
One remaining question is whether there are demographic or political cleavages in
attitudes toward criminal record checks in hiring. Again, prior research on attitudes toward
reentry has found few demographic cleavages (Garland et al., 2015, 2016). But, given that the
prevalence of criminal records varies across racial, gender, and socioeconomic groups (Pager,
2007; Western, 2006), it would not be surprising if these groups differed in their support for
allowing criminal record inquiries, particularly at the application stage where such inquiries
11 The BTB state variable is measured as whether the state had implemented a BTB policy at the time of the survey,
not at the time of the person’s last application. The results are the same if the BTB variable is removed.
might prevent applicants from ever getting their foot in the door. As shown above, criminal
record inquiries almost always happen at the application stage. Additionally, most of the
variation in public attitudes is between allowing criminal record inquiries at the application stage
versus other stages. The debate about BTB policies also centers on the application stage,
potentially because application-based criminal record inquiries serve a unique “gatekeeper
function” (Vuolo et al., 2017). As Vuolo and colleagues (2017, p. 1-2) note, “In theory, law, and
policy, it is important to distinguish between the initial application stage of the process and later
stages at which criminal background checks are routinely conducted.” For all of these reasons,
our analysis focuses on attitudes toward application-stage inquiries. We coded respondents “1” if
they supported allowing criminal record inquiries at the application stage, and “0” if they did not.
Inspection of Table 3 reveals very few correlates of support for criminal record questions
on job applications. There are no significant racial or gender differences in such attitudes, nor do
these attitudes vary significantly by education or income. Perhaps most notably, respondents
living in BTB and non-BTB states are equally likely to support allowing application-stage
inquiries. The only significant associations that emerge are for homeownership status, political
identification, and region of residence. Homeowners, Republicans, and Midwesterners are all
more likely to support application-stage inquiries, rather than delaying or banning inquires.
Homeownership reduces mobility (DiPasquale & Glaeser, 1999) and is strongly and positively
correlated with social capital (Glaeser, Laibson & Sacerdote, 2002). Therefore, our finding may
reflect the notion that homeowners are more invested in the community, and support for criminal
record inquiries may indicate a form of social control. The positive association between
Republicanism and support for criminal record inquiries is expected given that persons with
more conservative political ideologies tend to hold more negative attitudes toward people with
criminal records (Hirschfield & Piquero, 2010) and be less supportive of reentry initiatives
(Garland et al., 2015, 2016). However, even among Republicans there is considerable support for
waiting until after the application stage to ask about criminal records. More than 40% of “not
strong” Republicans, and 30% of “strong” Republicans, prefer either delaying or eliminating
criminal record inquiries.
Table 3 about here
Supplementary Analyses
One limitation of the analysis of experiences with hiring inquires (Table 2) is the
temporal ordering of the time of the last job search and respondents’ current characteristics.
Specifically, we were unable to account for whether respondents lived in a BTB state during
their last job search. We explore whether this influences the results in a set of sensitivity analyses
(see online supplements). In Model 1 of Supplement B, we adjust the BTB state measure based
on state adoption dates from the National Employment Law Project (Rodriguez & Avery, 2016).
Although the analysis is imperfect12, and we must use ranges for the time of the last job search,
we are able to estimate when respondents were not in active BTB states. We find that some of the
time since last job search measures are no longer statistically significant13, but the results
otherwise support the original results in Table 2, and the coefficients for male, age, education,
Midwest, and West are similar in magnitude.
Similarly, respondents may have changed education, income, homeownership, martial
and childcare statuses since they last applied for a job. In Model 2 of Supplement B, we restrict
12 We anticipate a small portion of our sample moved from one state to another between the last job application and
the time of the survey. Out of a population of around 315.1 million people in 2013, the U.S. Census estimates about
36 million people moved between 2012 and 2013, and less than a third of the movers (11.7 million) relocated to a
different county (which includes both intrastate and interstate moves) (Ihrke, 2014). This amounts to less than 4% of
U.S. residents ages 1 or older estimated to change counties within a year, with an even lower estimate for the
proportion of interstate (and inter-BTB policy) moves for adult who are actively in the labor market.
13 This is most likely a result of the necessity of collapsing the “> 10 years” and “5 to 10 years” response categories,
because too few of the respondents who last applied for a job 10 years ago lived in an active BTB state.
the analysis to applicants who report applying for their last job within the past two years (N=302)
to reduce the potential for major life changes. While age and education are no longer statistically
significant, due to the smaller sample size, the magnitudes are similar to the results in Table 2,
and homeowners are now much more likely to be asked about a criminal record on an application
than non-homeowners. The restricted sample in Model 2 is notably younger (36.7 on average,
which is about a decade younger than the sample average), and the results may represent a
different facet of the types of jobs to which these individuals apply. Still, being male and residing
in the Midwest are still strongly correlated (and in the same direction) with being asked about the
criminal record on the application—two of the most stable factors in the model.
In our analysis of attitudes toward criminal record inquiries (Table 3), one possible
correlate of support, as suggested by anonymous reviewers, is the respondent’s personal
experiences with the criminal justice system. Personal—and even vicarious—experiences can
influence respondents’ attitudes about informal social controls, such as neighbors intervening on
others’ behalf (Rose & Clear, 2004) and formal social controls, such as criminal justice
sentencing punitiveness (Davila, Hartley, Buckler, & Wilson, 2011). Mancini, Barrick, DiPonio,
and Gertz (2010) also find that prior experience with the criminal justice system is related to
increased fiscal support for correctional programs, and it is possible that this extends to
perceptions of criminal record inquiries for employment.
A limitation of the GfK Knowledge Networks survey is that due to resource constraints,
we could only include a limited set of questions, and we did not ask respondents about having a
criminal record. However, our research team also launched a more comprehensive survey about
criminal background checks within a few week timespan using SurveyMonkey, and
approximately 15% of SurveyMonkey respondents reported a prior arrest.14 The results in
Supplement C indicate respondents’ criminal record status is not an influential factor in their
support for employers asking about a job applicant’s criminal record. The other measures have
similar relationships with support in both the SurveyMonkey and GfK samples, with the
exception of the Midwest region. In both surveys, political ideology is the strongest predictor of
support. The SurveyMonkey data serve as a useful replication of the GfK findings using a
different national sample of U.S. adults.
As mentioned earlier, we specified our dependent variable of interest in the support
models as “on the application” versus all other categories to reflect the current BTB policy issue.
However, as a robustness check to the main results, we also remove respondents indicating an
employer should “never” ask about a criminal record and retain only those who believe
employers should ask at some point in the process. These findings are reported in the
Supplement D for both the GfK survey and the SurveyMonkey survey. Republicanism and
Midwest have a similar magnitude and statistical significance, and lower educational attainment
is now correlated with support for asking on the application relative to later in the process. As
before, prior arrest is not significantly related to inquiries on the application instead of later.
Finally, we include an analysis (see Supplement E) that compares “never” asking about a
criminal record to asking at some point in both the GfK and SurveyMonkey data. The results in
Supplement E indicate men are more likely than women to support never inquiring about a
criminal record, Republicans/Conservatives are less likely than Democrats/Liberals, and (in the
SurveyMonkey data) having a prior arrest is meaningfully and positively associated with support
for never inquiring. In sum, political ideology is the strongest indicator of support for criminal
14 See Supplement C for details on the SurveyMonkey methodology, sample descriptive statistics, and a logistic
regression predicting public support for allowing criminal record questions on job applications with prior arrest
experience included.
record inquiries on the application stage (compared to later/never), and for inquiring at any point
(relative to never). By contrast, having a prior arrest is related to respondents selecting “never”
relative to some type of inquiry, but it does not predict preferences for when the inquiry occurs in
the hiring process (i.e., the application stage versus later/never).
Discussion and Policy Implications
Recent Ban-the-Box (BTB) policy discussions have targeted the application stage of the
hiring process and enacted legislation to prohibit employers from asking about criminal records
until later stages (e.g., interview or job offer) in the hiring process. The key assumptions
underlying the BTB policy approach are that criminal record inquiries are common, and tend to
occur at the application stage. To examine the validity of these assumptions, and incorporate
public opinion research into the debate, we analyzed data from a probability sample of U.S.
residents. Several important findings emerged from our analysis, and we discuss each below.
First, we provide the first nationally representative estimates of U.S. adults’ experiences
with criminal record inquiries in the job hiring process. The survey reveals an important fact: the
majority of U.S. residents (58%) who apply for a job are asked about their criminal record at
some point in the process, and this almost always occurs on the job application. Furthermore,
respondents applying for jobs in recent years had estimates closer to 70%. If anything, these
figures are likely underestimates due to recall failure. To place the personal experience results
into the context of the existing literature, prior estimates include: 63% of employers conducting
checks (Holzer, Raphael & Stoll, 2006), 78% of employers asking on job applications (Vuolo et
al., 2017), and 87% of employers conducting checks on some or all job candidates (Society for
Human Resource Management, 2012). Taken together, the evidence to date suggests that
criminal record evaluations are now ubiquitous in hiring, whether by inquiries to applicants or
external checks.
Second, we find that virtually all of the action for criminal record inquiries in hiring is at
the application stage. Specifically, the findings show that among respondents who were asked by
employers about their criminal records, approximately 87% were first asked on the application
form. Indeed, our data suggest that over 31 million U.S. adults were asked about their criminal
record on a job application form in just the past year. This clearly justifies concerns about
potential “close-the-door” negative effects of early criminal record inquiries, and gives credence
both to BTB’s focus on application forms and scholarship studying the nature of application-
stage checks (Vuolo et al., 2017). When considering that 16% of U.S. residents are convicted of a
crime by age 23 (Bushway, Cutler, Paternoster, & Brame, 2014), application-stage criminal
record inquiries clearly have the potential to exert a major impact on the nature of the U.S. labor
Third, our results show that the public is roughly split in its support for application-stage
inquiries. The vast majority (91%) of respondents support allowing employers to ask job
applicants about their criminal records, but only a slight majority (57%) prefers allowing
employers to include this question on application forms. Respondents who reside in BTB states
are just as likely as those who do not to support criminal record inquiries on the job application.
Therefore, the findings suggest that policymakers would be acting consistently with popular
attitudes by allowing the inclusion of criminal record questions on job application forms. This is
important because the BTB policy option might warrant serious reconsideration if evidence
demonstrating statistical discrimination under BTB continues to accumulate (Agan & Starr,
2016; Doleac & Hansen, 2016; Vuolo et al., 2017).
Our findings also suggest, however, that if evidence accumulates that delaying criminal
record inquiries has a net benefit, many members of the public would be open to postponing
these inquiries. Even among Republicans, who are significantly more likely than others to
support application-stage inquiries, there is already a sizable minority (roughly 35%) who prefers
delaying or eliminating criminal record inquiries. In this sense, the “boundaries of policy
initiatives” (Thielo et al., 2015, p. 141) currently set by public opinion for regulating criminal
record inquiries in hiring appear to be very wide. This would seemingly suggest that
policymakers have great latitude to develop evidence-based policies for employers’ use of
criminal records without having to fear public backlash or electoral sanctions.
One unexpected finding was that homeowners were significantly more likely to support
application-stage criminal record inquiries, relative to delaying or banning inquiries. Indeed, the
odds of support for application-stage inquiries were 78% higher for homeowners than those who
did not own their homes. One possible explanation is the “not-in-my-backyard (NIMBY)”
phenomena.15 It may be that homeowners perceive that application-stage checks, by limiting
certain employment opportunities to those with criminal records, or at least to those with
particular types of records (e.g., violent), help to create social distance between persons with
criminal records and those without records. Most people recognize that buying a home in an
owner-dominated residential area requires considerable financial resources and stable
employment. It may be that homeowners believe that restricting job access through application-
stage criminal record inquiries represents an initial line of defense for protecting their
communities from outsiders with records. Future research is needed that explores this possibility.
15 NIMBY was originally conceptualized to help city planners understand opposition to the development of certain
types of facilities, such as low-income housing, hazard waste storage, or social services facilities (Dear, 1992).
While residents accept that the facilities are necessary, they also do not want them located near their homes (Dear,
There are several limitations to this study. First, given the low survey response rate, there
remains the possibility of nonresponse bias in our findings. Even still, our data provide the best
available estimates of public support for and experience with background checks, deriving from
a self-administered survey recently administered to a random sample of Americans by one of the
most highly respected and widely used survey firms (The GfK Group).16 Nonetheless, future
studies are needed to replicate our results using surveys with higher response rates. Second, with
self-reports over extended reference periods, recall errors are always a consideration. It is
possible respondents incorrectly recalled when their last job search occurred or whether or when
they were asked about their criminal record. Researchers seeking to build on our analysis might
consider using longitudinal studies with bounded interviews and shorter reference periods to
examine experiences with criminal record inquiries in hiring. Our data reveal a relatively high
incidence of job searches—over 21% of U.S. adults applied for a job in just the past year. Thus,
there would appear to be sufficient variation in job searches to make a longitudinal investigation,
perhaps over 2-3 years, worthwhile. Similarly, incorporating additional detail into the experience
question, such as distinguishing between job searches that occurred when respondents were
employed but looking elsewhere versus unemployed, or exploring variations by industry and job
type, would also be useful to explore in future research.
Not least, national probability samples are advantageous for obtaining reliable estimates
of experiences and perceptions, but it is not typically financially feasible to include a large set of
auxiliary questions with which to test competing theoretical models of attitudes. Future efforts
could utilize non-probability samples to further theory-test and investigate correlates of attitudes
toward criminal background inquiries and experiences. For example, studies might explore
16 The GfK Group’s KnowledgePanel has been used frequently to field general population surveys for the American
National Election Studies, and for the Time-Sharing Experiments for the Social Sciences, an ongoing
interdisciplinary program funded by the National Science Foundation (NSF) (Mutz, 2011).
whether various measures of contact with persons convicted of crimes are correlated with views
about criminal record inquiries in hiring (see, e.g., Hirschfield & Piquero, 2010).
We conclude by emphasizing that historical trend data indicate that in the foreseeable future, well
over half a million inmates will be released annually from state and federal prisons (Carson,
2015). The majority of these individuals will eventually submit a job application (Travis, 2005),
the success of which has implications for their likelihood of transitioning out of a life of crime
(Laub & Sampson, 2001; Uggen, 2000). Our findings strongly suggest that most of them will be
asked about their criminal record very early in the hiring process, before their other
qualifications are considered. Prior research reveals that for many employers, a criminal record is
a nonstarter—they simply do not want to hire individuals who have been convicted of a crime
(Pager, 2007). For this reason, continuing efforts to document when and how criminal records
are disclosed and used during the hiring process are important. More generally, ensuring that
applicants who have criminal records receive fair consideration in hiring decisions is a pressing
policy issue.
Agan, A., & Starr, S. (2016). Ban the Box, criminal records, and statistical discrimination: A field
experiment. Law and Economics Research Paper Series. Paper No. 16-012.
Alexander, M. (2012). The new Jim Crow: Mass incarceration in the age of colorblindness. The
New Press.
Allcott, H. (2011). Consumers’ perceptions and misperceptions of energy costs. American
Economic Review, 101, 98-104.
Amaya, A., & Presser, S. (2017). Nonresponse bias for univariate and multivariate estimates of
social activities and roles. Public Opinion Quarterly 81, 1-36.
American Association for Public Opinion Research. (2016). Standard definitions: Final
dispositions of case codes and outcome rates for surveys. Ann Arbor, MI: American
Association for Public Opinion Research.
Blumer, Herbert. 1969. Symbolic Interactionism: Perspective and Method. Berkley, CA:
University of California Press.
Brame, R, Bushway, S. D., Paternoster, R., & Turner, M. G. (2014). Demographic patterns of
cumulative arrest prevalence by ages 18 and 23. Crime & Delinquency, 60: 471–86.
Brame, R., Turner, M. G., Paternoster, R., & Bushway, S. D. (2012). Cumulative prevalence of
arrest from ages 8 to 23 in a national sample. Pediatrics, 129, 21–27.
Bureau of Justice Statistics. (2001). Privacy, technology and criminal justice information: Public
attitudes toward uses of criminal history information. U.S. Department of Justice, Office
of Justice Programs, Bureau of Justice Statistics.
Bureau of Labor Statistics. (2015). Table 11: Employed persons by detailed occupation, sex,
race, and Hispanic or Latino ethnicity. Labor Force Statistics from the Current Population
Survey. Retrieved July 25, 2016 from http://www.bls.gov/cps/cpsaat11.htm.
Bushway, S. D. (2004). Labor market effects of permitting employer access to criminal history
records. Journal of Contemporary Criminal Justice, 20, 276–91.
Bushway, S. D., Cutler, C., Paternoster, R., & Brame, R. (2014). Demographic patterns of
cumulative conviction prevalence to age 26. Presentation to Stockholm Symposium, June
Carson, A. E. (2015). Prisoners in 2014. Washington, DC: Bureau of Justice Statistics,
U.S. Department of Justice.
Canes-Wrone, B., & Shotts, K. W. (2004). The conditional nature of presidential responsiveness
to public opinion. American Journal of Political Science, 48, 690-706.
Chang, L., & Krosnick, J. A. (2009). National surveys via RDD telephone
interviewing versus the internet: Comparing sample representativeness and response
quality. Public Opinion Quarterly, 73, 641-678.
Davila, M. A., Hartley, D. J., Buckler, K., & Wilson, S. (2011). Personal and vicarious
experience with the criminal justice system as a predictor of punitive sentencing
attitudes. American Journal of Criminal Justice, 36, 408-420.
Dear, M. (1992). Understanding and overcoming the NIMBY syndrome. Journal of the
American Planning Association, 58, 288-300.
DiPasquale, D. & Glaeser, E. (1999). Incentives and social capital: Are homeowners better
citizens? Journal of Urban Economics, 45, 354–84.
Doleac, J. L., & Hansen, B. (2016). Does ‘Ban the Box’ help or hurt low-skilled workers?
Statistical discrimination and employment outcomes when criminal histories are hidden.
Presentation to the Institute for Research on Poverty at the University of Wisconsin-
Madison, June 22.
Enns, P. K. (2016). Incarceration nation: How the United States became the most punitive
democracy in the world. New York: Cambridge University Press.
Equal Employment Opportunity Commission. (2012). EEOC enforcement guidance:
Consideration of arrest and conviction records in employment decisions under Title VII
of the Civil Rights Act of 1964. Washington, DC: U.S. EEOC.
Ewald, A. (2016). Licensing barbers: Street-level bureaucrats and discretionary citizenship in the
expanded American carceral state. Working paper.
Federal Interagency Reentry Council. (2016). The Federal Interagency Reentry Council: A
Record of Progress and a Roadmap for the Future. Washington DC. The White House
and U.S. Department of Justice.
Furstenberg, F. F., Kennedy, S., McLoyd, V. C., Rumbaut, R. G., & Settersten, R. A. (2004).
Growing up is harder to do. Contexts, 3, 33–41.
Garland, B., Wodahl, E., & Cota, L. (2016). Measuring support for prisoner reentry options.
International Journal of Offender Therapy and Comparative Criminology 60, 1406-
Garland, B., Wodahl, E., & Saxon, C. (2017). What influences public support of transitional
housing facilities for offenders during reentry? Criminal Justice Policy Review, 28, 18-
Garland, B., Wodahl, E. & Schuhmann, R. (2013). Value conflict and public opinion toward
prisoner reentry initiatives. Criminal Justice Policy Review, 24, 27-48.
Garland, B., Wodahl, E., & Smith, R. G. (2015). Religious beliefs and public support for prisoner
reentry. Criminal Justice Policy Review, OnlineFirst. doi 0887403415619985.
Glaeser, E. L., Laibson, D., & Sacerdote, B. (2002). An economic approach to social capital. The
Economic Journal, 112, F437-F458.
Goffman, Erving. 1963. Stigma: Notes on the Management of Spoiled Identity. London: Penguin
Groves, R. M. (2006). Nonresponse rates and nonresponse bias in household surveys.
Public Opinion Quarterly, 70, 646–75.
Groves, R. M., & Peytcheva, E. (2008). The impact of nonresponse rates on nonresponse bias.
Public Opinion Quarterly, 72, 167-189.
Groves, R. M., Fowler, F. J., Jr., Couper, M. P., Lepkowski, J. M., Singer, E., & Tourangeau, R.
(2009). Survey Methodology (2nd ed.). Hoboken, NY: Wiley.
Heggestad, E. D., Rogelberg, S., Goh, A., & Oswald, F. L. (2015). Considering the effects of
nonresponse on correlations between survey variables: A simulation study to provide
context to evaluate survey results. Journal of Personnel Psychology 14:91-103.
Hirschfield, P. J., & Piquero, A. R. (2010). Normalization and legitimation: Modeling
stigmatizing attitudes toward ex-offenders. Criminology, 48, 27-55.
Holzer, H. J., Raphael, S. & Stoll, M. A. (2002). Will employers hire ex-offenders? Employer
preferences, background checks and their determinants. In The impact of incarceration
on families and communities. Edited by Mary Pattillo, Bruce Western and David Weiman.
New York: Russell Sage Foundation.
Holzer, H. J., Raphael, S. & Stoll, M. A. (2006). Perceived criminality, criminal background
checks, and the racial hiring practices of employers. Journal of Law and Economics, 49,
Ihrke, David. (2014). Reason for Moving: 2012 to 2013. Population Characteristics. Washington,
DC: United States Census Bureau: U.S. Department of Commerce.
Irwin, J. (1970). The Felon. Berkeley, CA: University of California Press.
Jacobs, J. B. (2015). The eternal criminal record. Harvard University Press.
Kaeble, D., Glaze, L., Tsoutis, A., & Minton, T. (2015). Correctional populations in the United
States, 2014. U.S. Department of Justice, Office of Justice Programs, Bureau of Justice
Kingdon, J. W. 1995. Agendas, Alternatives and Public Policies. 2nd ed. New York: Harper
Krosnick, J. A., Presser, S., Fealing, K. H., & Ruggles, S. (2015). The Future of Survey
Research: Challenges and Opportunities. The National Science Foundation Advisory
Committee for the Social, Behavioral and Economic Sciences Subcommittee on
Advancing SBE Survey Research. Available online at:
Lageson, S., Vuolo, M., & Uggen, C. (2015). Legal ambiguity in managerial assessments of
criminal records. Law and Social Inquiry, 40, 175–204.
Laub, J. H., & Sampson, R. J. (2001). Understanding desistance from crime. Crime and Justice,
28, 1–41.
Lemert, E. (1951). Social Pathology. New York: McGraw-Hill.
Levin, M. (2015). Koch Industries to stop asking about job candidates’ criminal history. Politico.
Retrieved January 15, 2017 from: http://www.politico.com/story/2015/04/koch-
Mancini, C., Barrick, K., DiPonio, J., & Gertz, M. (2010). Taxation with representation?
Examining public fiscal support for diverse correctional policies. Criminal Justice Policy
Review, 21, 76-97.
Mead, G. H. (1934). Mind, Self and Society. Vol. 111. Chicago: University of Chicago Press.
Mullainathan, S. (2016). Ban the Box? An effort to stop discrimination may actually increase it.
New York Times. August 19th, 2016.
Mutz, D. C. (2011). Population-based Survey Experiments. New Jersey: Princeton University
Nagin, D. S., Piquero, A. R., Scott, E. S., & Steinberg, L. (2006). Public preferences for
rehabilitation versus incarceration of juvenile offenders: Evidence from a contingent
valuation survey. Criminology & Public Policy, 5, 627-651.
Nicholson-Crotty, S., Peterson, D. A. M., & Ramirez, M. D. (2009). Dynamic representation(s):
Federal criminal justice policy and an alternative dimension of public mood. Political
Behavior, 31, 629-655.
Ouellette, H., Applegate, B., & Vuk, M. (2016). The public’s stance on prisoner reentry: Policy
support and person acceptance. American Journal of Criminal Justice (published online
December 2016).
Pager, D. (2003). The mark of a criminal record. American Journal of Sociology, 108, 937–75.
Pager, D. (2007). Marked: Race, crime, and finding work in an era of mass incarceration.
Chicago, IL: University of Chicago Press.
Pager, D., & Quillian, L. (2005). Walking the talk: What employers say versus what they do.
American Sociological Review, 70, 355–80.
Petersilia, J. (2003). When Prisoners Come Home: Parole and Prisoner Reentry. New York:
Oxford University Press.
Reno, J. (2000). Remarks of the Honorable Janet Reno Attorney General of the United States on
Reentry Court Initiative. United States Department of Justice. Retrieved January 25,
2017 from https://www.justice.gov/archive/ag/speeches/2000/doc2.htm
Rodriguez, M. N., & Avery, B. (2016). Ban the Box: U.S. cities, counties, and states adopt fair
hiring policies. National Employment Law Project. Retrieved July 25, 2016 from
Rogelberg, S. G., & Stanton, J. M. (2007). Introduction: Understanding and dealing with
organizational survey nonresponse. Organizational Research Methods 10, 195-209.
Rose, D. R., & Clear, T. R. (2004). Who doesn’t know someone in jail? The impact of exposure
to prison on attitudes toward formal and informal controls. The Prison Journal 84, 228-
SEARCH. (2005). Report of the national task force on the commercial sale of criminal justice
record information. Sacramento, Calif.: SEARCH Group, Inc.
Semuels, A. (2016). When banning one kind of discrimination results in another. The Atlantic.
August 4, 2016.
Siwach, G. (2016). Criminal background checks and recidivism: Bounding the causal impact.
Presented at the American Society of Criminology, November 16.
Society for Human Resources Management. (2012). Background checking: The use of criminal
background checks in hiring decisions. Retrieved July 25, 2016 from
Solove, D. J., & Hoofnagle, C. J. (2006). A model regime of privacy protection (version
3.0). University of Illinois Law Review, 2, 357-403.
Shoag, D., & Veuger, S. (2016). No woman no crime: Ban the Box, employment, and upskilling.
Working paper.
Tannenbaum, F. (1938). Crime and the Community. Boston: Ginn.
The Council of State Governments. (2017). The National Inventory of Collateral Consequences
of Conviction. Last accessed January 19, 2017: https://niccc.csgjusticecenter.org/map/
The Southern Coalition for Social Justice. (2014). The benefits of Ban the Box: A case study of
Durham, NC.
Thielo, A. J., Cullen, F. T., Cohen, D. M., & Chouhy, C. (2015). Rehabilitation in a red
state. Criminology & Public Policy, 15, 137-170.
Toney, M.W. (2007). A Second Chance—for the First Time: Movement Formation Among
Formerly Incarcerated People. Berkeley, CA: University of California Press.
Tourangeau, R., Maitland, A., & Yan, H. Y. (2016). Assessing the scientific knowledge of the
general public: The effects of question format and encouraging or discouraging don’t
know responses. Public Opinion Quarterly, Advanced access. doi: 10.1093/poq/nfw009.
Travis, J., Western, B., & Redburn, S. (2014). The growth of incarceration in the United States:
Exploring causes and consequences. Washington, DC: National Academies Press.
Travis, J. (2005). But they all come back: Facing the challenges of prisoner reentry. Washington,
D.C.: The Urban Institute Press.
Uggen, C. (2000). Work as a turning point in the life course of criminals: A duration model of
age, employment, and recidivism. American Sociological Review, 67, 529-46.
Uggen, C., Vuolo, M., Lageson, S., Ruhland, E., & Whitham, H. K. (2014). The edge of stigma:
An experimental audit of the effects of low-level criminal records on employment.
Criminology, 52, 627-654.
Vuolo, M., Lageson, S., & Uggen, C. (2017). Criminal record questions in the era of
‘Ban the Box.’ Criminology & Public Policy. In press.
Watstein, R. D. (2009). Out of jail and out of luck: The effect of negligent hiring liability and the
criminal record revolution on an ex-offender's employment prospects. Florida Law
Review, 61, 581-609.
Weinberg, J. D., Freese, J., & McElhattan, D. (2014). Comparing data characteristics and results
of an online factorial survey between a population-based and a crowdsource-recruited
sample. Sociological Science, 1, 292–310.
Western, B. (2006). Punishment and inequality in America. Russell Sage Foundation.
White House. (2015). FACT SHEET: President Obama announces new actions to promote
rehabilitation and reintegration for the formerly-incarcerated. Washington, DC: The
White House, Office of the Press Secretary.
Zahariadis, N. (2014). Ambiguity and multiple streams. In Paul A. Sabatier and Christopher
Weible (eds.), Theories of the Policy Process. Boulder, CO: Westview Press.
Table 1 Prevalence of criminal record questions in respondents’ most recent job search, by
timing of last application
Sample On Application Form
During Interview
After Hiring Decision
Never Asked
Total sample (N = 983) 51 4 3 42
Time since application
< 1 year (n = 216) 59 8 4 29
1 to 2 years (n = 96) 63 2a 6a30
3 to 4 years (n = 114) 64 7a 4a26
5 to 10 years (n = 180) 54 3a 3a40
> 10 years (n = 377) 38 3 2a57
Notes. Numbers do not sum to 100 because of rounding. Respondents (n = 19) who said they had never applied for a
job were excluded. aEstimate is based on fewer than 10 cases.
Table 2 Logistic regression predicting whether respondents were asked about their criminal
record on the application form during their most recent job search (N = 962)
95% Confidence Interval
Variables bSE Lower Upper
Time of Last Job Search
> 10 years (reference)
5 to 10 years .428* (.217) .002 .855
3 to 4 years .852** (.268) .326 1.379
1 to 2 years .634* (.278) .088 1.180
< 1 year .361 (.238) –.105 .827
Black .249 (.278) –.296 .794
Latino –.220 (.259) –.727 .288
Male –.371* (.149) –.664 –.078
Age –.018** (.006) –.029 –.006
Education .175** (.053) .071 .278
Income –.002 (.059) –.118 .114
Unemployed –.169 (.326) –.810 .471
Homeowner .255 (.185) –.108 .619
Married –.192 (.173) –.531 .148
Youth (< 18) in household .229 (.190) –.144 .602
Republicanism .003 (.037) –.069 .075
Northeast (reference)
Midwest .469* (.233) .013 .926
South .144 (.226) –.300 .588
West .503* (.237) .039 .968
Resides in a BTB State .009 (.165) –.315 .334
Notes. Outcome is coded: 1 = asked about criminal record on application form in last job search, 0 = not asked or
asked at later stage.
*p < .05; **p < .01; ***p < .001 (two-tailed).
Figure 1 Respondents’ preferences for when employers should first be allowed to ask job applicants
about their criminal records (N = 1,009)
Table 3 Logistic regression predicting public support for allowing criminal record questions on
job applications (N = 987)
95% Confidence Interval
Variables bSE Lower Upper
Black .053 (.257) –.452 .558
Latino .168 (.249) –.321 .657
Male –.016 (.145) –.299 .268
Age .004 (.005) –.006 .014
Education –.057 (.051) –.157 .043
Income .013 (.058) –.102 .127
Unemployed –.244 (.314) –.859 .372
Homeowner .575** (.176) .229 .921
Married –.121 (.169) –.453 .211
Youth (< 18) in household .011 (.181) –.345 .367
Republicanism .114** (.036) .043 .185
Northeast (reference)
Midwest .452* (.214) .032 .872
South .355 (.207) –.051 .762
West .300 (.227) –.146 .746
Resides in a BTB State .191 (.155) –.114 .496
Notes. Outcome is coded: 1 = on the application form; 0 = at the interview stage, after hiring decision, or never.
*p < .05; **p < .01; ***p < .001 (two-tailed).
... Worldwide, employment recruiters are increasingly conducting criminal background checks (CBCs), seeking to gather information on previous criminal convictions held by applicants prior to the hiring decision (Denver et al., 2018;Larrauri, 2014). 1 This may have unfortunate consequences because, when CBCs are prevalent in the labour market, the chances of being hired are low for people with previous convictions (Bushway, 2004). This is because employers do not always base their hiring decisions on a rational risk assessment when they know that a job applicant has been convicted in the past (Rodríguez Menés and Rovira, 2019;Sugie et al., 2020). ...
... In addition to technological facilitation, the growth in the use of CBCs has been fuelled by changes in societal values, which have made employers more willing to request information on previous convictions from job applicants (Holzer et al., 2007). These include increased public support for penal control measures aimed at the social exclusion of sexual offenders (McAlinden, 2012;Thomas and Hebenton, 2013) and increased public acceptance of a role for private actors in crime prevention (Denver et al., 2018;Garland, 2001). ...
Full-text available
The use of criminal background checks (CBCs) – requests for information on previous convictions during the employment recruitment process – is growing worldwide. This article seeks to explain the proliferation in CBCs by examining whether novel legislation introducing mandatory requests for some jobs also leads to an increase in requests for CBCs for jobs outside the scope of the law. The present research makes use of survey data collected from individuals requesting criminal records certificates before and after the introduction of new CBC regulation in Spain – EU Directive 93/2011/EU – which established the obligation to request a criminal record certificate covering sexual crimes for jobs involving frequent contact with children. The analysis detects only a small and unsustained growth in non-mandatory checks following introduction of the new law. However, the results suggest that the danger of the new legislation lies in employers requesting certificates with a higher level of disclosure than is required for the positions on which checks were made mandatory by the new law. In addition, the growth in non-mandatory CBCs observed during this period seems to be related not to the new legislation but to the emergence of tech companies, raising alarm regarding the role of novel forms of policy mobility and the new collaborative economy in limiting the re-entry of individuals with criminal records to the labour market.
... We encourage researchers to abandon the use of "agree-disagree" (AD) survey items whenever possible because they are problematic no matter the order. Instead, we concur with Pickett and Baker's (2014, p. 215) recommendation to consider "item-specific" (IS) questions, which feature tailored response options (see, e.g., Denver et al., 2018). Evidence suggests the use of IS questions results in less acquiescence bias (Saris et al., 2010), more conscientious responding (Höhne et al., 2018), and overall better data quality (for a review, see Dykema et al., 2021). ...
Full-text available
In 2014, Pickett and Baker cast doubt on the scholarly consensus that Americans are pragmatic about criminal justice. Previous research suggested this pragmatism was evidenced by either null or positive relationships between seemingly opposite items (i.e., between dispositional and situational crime attributions and between punitiveness and rehabilitative policy support). Pickett and Baker (2014) argued that because these studies worded survey items in the same positive direction, respondents’ susceptibility to acquiescence bias led to artificially inflated positive correlations. Using a simple split-ballot experiment, they manipulated the direction of survey items and demonstrated bidirectional survey items resulted in negative relationships between attributions and between support for punitive and rehabilitative policies. We replicated Pickett and Baker’s (2014) methodology with a nationally representative sample of American respondents supplemented by a diverse student sample. Our results were generally consistent, and, in many cases, effect sizes were stronger than those observed in the original study. Americans appear much less pragmatic when survey items are bidirectional. Yet, we suggest the use of bidirectional over unidirectional survey items trades one set of problems for another. Instead, to reduce acquiescence bias and improve overall data quality, we encourage researchers to adopt item-specific questioning.
... In terms of race, findings suggest that compared to non-Hispanic Whites, Hispanics, and non-Hispanic Blacks are more accepting of incarceration histories (Hirschfield & Piquero, 2010), while others report that Asians and Hispanics are more punitive (Socia et al., 2019). Those with more conservative beliefs consistently hold more negative views towards CFIP (Denver et al., 2018;Rade et al., 2016). Higher levels of education are associated with increased support for rehabilitation (Baker et al., 2015;Baker et al., 2016), but this, too, is not always the case (Hirschfield & Piquero, 2010). ...
We examine how public attitudes toward currently/formerly incarcerated people and their reentry into society are affected by consuming information about imprisoned people. Over 1,500 respondents from a national online survey were randomly assigned one of five sources of written information about currently and formerly incarcerated people (CFIP) (three informative pieces and two sets of incarceree poetry). They then reported their views toward them and support for reentry policies. While no differences in support for reentry initiatives across conditions were uncovered, those reading poetry with a humanizing theme applied the least amount of stigma toward currently/formerly incarcerated people.
... First, in a 2016 survey (N = 1,009), Denver et al. (2018) asked the following question: "In your view, when should employers FIRST be allowed to ask about a job applicant's criminal record, or do you think they should never be allowed to ask?" Almost 6 in 10 respondents (57 percent) chose "on the job application, " suggesting opposition to the ban-the-box policy. The other results were 28 percent for "at the interview stage, " 6 percent for "after the hiring decision, " and 9 percent for "never. ...
Full-text available
In The New Jim Crow, Michelle Alexander drew national attention to the extensive imposition of collateral consequences on those convicted of a crime and to their racially disparate effects. These so-called invisible punishments are intended to restrict offender participation in civil, economic, and social institutions. At issue, however, is how the American public views collateral consequences and whether it would support a reduction in their usage. Based on a 2017 national- level YouGov survey (N = 1,000), supplemented by a second 2019 YouGov survey (N = 1,200), the current study finds that the public is split on allowing ex-offenders to sit on juries, but supportive of removing barriers to voting and employment (i.e., endorses ban-the-box reforms). The respondents also favored providing defendants with a list of restrictions linked to conviction as well as having lawmakers review and eliminate collateral consequences found to have no purpose and to not reduce crime. Belief in offender redeemability is a robust source of these policy preferences. Overall, the American public, including White adults, seems prepared to move beyond “the new Jim Crow” by reforming a legacy of the get-tough era—a byzantine system of collateral consequences that imposes disabilities without transparency and with no evidence of their efficacy.
... First, the current study uses a convenience sample. Although an increasing number of studies that use nonprobability samples have appeared in top criminology journals (e.g., Enns and Ramirez 2018;Denver et al. 2018;Vaughan et al. 2019), Thompson and Pickett (2019) have observed that crowdsource and opt-in panels are not equivalent to random sampling schemes in terms of generalizability. Future researchers should address this limitation by examining the research questions using a nationally representative sample. ...
Full-text available
Objectives: Public punitiveness is closely related to the expansion of the U.S. penal system. Prior studies have examined inaccurate crime trend perceptions and negative emotions as key predictors of punitive attitudes. However, the interconnections between crime trend perceptions, negative emotions, and punitive attitudes have not been explored. It is yet unknown if exposure to accurate crime trend information reduces negative emotions and public punitiveness. Methods: I analyzed data from a survey-based experiment with a nationwide sample (N = 441) using conditional process analysis. Results: Perceptions of rising crime trends were related to punitiveness both directly and indirectly through anger about crime. Exposure to accurate crime information did not alleviate anger about or fear of crime, but, surprisingly, increased support for punitive criminal justice policies. Conclusions: The study highlights the importance of understanding the expressive and emotional elements of punishment, and the possible positive effect of providing people with accurate crime information on punitiveness.
Public opinion is doubly important for reintegration, as it shapes both the policy and the stigma environments that people with criminal records must face. Nowhere are the policy and stigma environments bleaker than for record holders convicted of sex crimes. Drawing on the theory of compassion collapse (or psychic numbing) and using experimental data from a national survey, we examine the effects of informing members of the public about the hardships faced by record holders convicted of sex crimes, and we compare those effects (or the lack thereof) to the effects of victim discourse. We also randomize the information format: aggregate/statistical versus personal narratives. We find that narratives about crime victims’ suffering matter to the public—increasing aversive emotions, support for collateral consequences, and stigmatization—but narratives about record holders’ suffering do not. We conclude by discussing alternative communication strategies that public criminologists may use to garner public support for progressive criminal justice reforms.
Obtaining employment is a major barrier to social reintegration for people on probation or parole. Research on the reentry process identifies several mechanisms that accentuate difficulties in locating work, including human capital development, structural changes in the labor market, and onerous probation and parole conditions. In this article, we review theories that explain low labor market participation rates among people reentering society, and we draw on multiple sources of data to identify the types of jobs that are available to people with low human capital. We find that nearly a quarter of people in America’s state and federal prisons had permanently removed themselves from the formal labor market before their most recent arrest; however, exclusionary hiring practices in the formal labor market often push those carrying the stigma of a criminal record into underground or informal labor markets, where wage rates are markedly higher than the federal minimum wage. Our findings demonstrate that severe and chronic employment struggles often predate and follow incarceration. We provide a detailed discussion of policy reform proposals that could help to remedy this harmful dynamic.
Regardless of why it happens, racial discrimination is damaging and unacceptable. Efforts to reduce discrimination, however, are most successful when we understand the mechanisms that give rise to it. Building on the observation that employers are members of the public, we examine two attitudinal mechanisms that may foster discriminatory employment practices in the context of criminal background checks: stereotypes and threat-based animus. First, we estimate public perceptions of arrest prevalence using two nationwide surveys. Next, we experimentally test the effects of two racially threatening primes—Census projections about a coming majority-minority America, and information about the prison population’s racial composition—on attitudes toward hiring job applicants with criminal records. Consistent with statistical discrimination theory, respondents identify black males as having the highest arrest prevalence. Respondents are less accurate, however, when it comes to gender differences: they underestimate arrest prevalence for black, Hispanic, and white males, and tend to overestimate it for females. On the other hand, our experiments provide little evidence of an effect of threat-based animus: racially threatening primes that are influential in other contexts do not significantly impact attitudes about hiring applicants with criminal records.
Over 11 million people in England and Wales have criminal records, with men, people from Black, Asian and Minority Ethnic backgrounds and people from poorer communities being over-represented. A gatekeeping discussion is developing across the social work profession regarding the role of criminal records in decision-making at the point of admissions to social work programmes. Balancing risk, representation, and the right to move on, who becomes a social worker is an issue of debate among practitioners, educators, and policy makers, where notable instances of media and political focus on social work failure has led to public mistrust, and moral panic about the state of the profession. Whilst internationally many social work education providers have moved away from interview gatekeeping, in England this has continued, alongside wider checks including for criminal records. This discussion article reviews what social working with a criminal record represents in the profession and why course admissions is a site for restorative justice and the human right to move on.
Security risk to information exists in many forms and guises; being human and non-human, and those directed from inside and outside of an organisation. Traditional mitigation methods have predominately focused upon applying layers of reactive security solutions, triggered by breaches to defined rules and processes. In order to counter the risk to information posed by individuals from within an organisation, behavioural research is increasingly becoming a crucial element in the design of automated security tools to identify and respond to anomalous and malicious behaviour. Accurate user profiles and threat models may enable the effective and efficient deployment of user behaviour analytics systems, which in turn enable the proactive mitigation of threats before they cause harm to people, property, including intellectual property, or premises. The consideration of personal characteristics and role-related behaviours during the design of security systems can protect the privacy rights of individuals, improve business effectiveness and cost efficiencies for organisations. In an increasingly mobile world, next generation and anticipatory security risk systems are needed to respond to the shift that people are becoming the new security perimeter.
Full-text available
Criminal background checks are increasingly being incorporated into hiring decisions by employers. Although originally uncompromising—almost anyone with a criminal record could be denied employment—court rulings and policy changes have forced criminal background checks to become more nuanced. One motivation for allowing more individuals with criminal records to work is to decrease recidivism and encourage desistance. In this article, we estimate the causal impact of receiving a clearance to work on subsequent arrests for individuals with criminal records who have been provisionally hired to work in certain nonlicensed health-care jobs in New York State (N = 6,648). We employ an instrumental variable approach based on a substantive understanding of the state-mandated criminal background check process. We examine age-graded effects within this group of motivated individuals and differential effects by sex in the rapidly growing health-care industry, which is typically dominated by women. Our estimated local average treatment effect indicates a 2.2-percentage-point decrease in the likelihood of a subsequent arrest in 1 year and a 4.2-percentage-point decrease over 3 years. We find meaningful variations by sex; men are 8.4 percentage points less likely to be arrested over the 3-year period when cleared compared with a 2.4-percentage-point (and nonsignificant) effect for women. Older women in particular are driving the nonsignificant results for women.
This paper estimates the effect of employment denial based on a criminal background check on recidivism outcomes for individuals with convictions who are provisionally hired in the New York State healthcare industry. Using institutional knowledge about the New York State Department of Health's screening process, I build structural assumptions on potential outcomes for different subsamples in my data, which partially identifies the Average Treatment Effects. I find a 0–2.2 percentage-point increase in the likelihood of subsequent arrests caused by employment denial, with substantial heterogeneity across the sample. Specifically, I find that the a priori highest risk individuals are most likely to be impacted by a loss of employment opportunity based on their criminal background. Policy implications of these results are discussed.
How does public opinion affect presidential policymaking? We address this issue by testing a diverse set of hypotheses with data concerning a set of individual policies across time. In particular, the data revolve around presidential budgetary proposals on a set of major policy issues for which there are recurring surveys on citizens' preferences over spending. The analysis suggests that presidents are more responsive to mass opinion on issues that are familiar to citizens in their everyday lives. Also, for reelection-seeking presidents, responsiveness is shown to depend upon two key political factors. First, presidents are more responsive to public opinion when the next election is imminent. Second, the effect of presidential popularity is nonmonotonic; presidents with average approval ratings are most likely to adopt policy positions congruent with public opinion, whereas presidents with approval ratings that are significantly above or below average have the greatest propensity to take unpopular positions.
Research Summary This study examines three central questions about criminal record inquiries on job applications, which is a rapidly developing area in criminology and public policy. We find the following: (1) Among the 78% of employers who ask about records, specific application questions vary greatly regarding the severity and timing of offenses. (2) Applications for restaurant positions are least likely to inquire about criminal histories, whereas racially diverse workplaces and establishments in the most and least advantaged neighborhoods are more likely to ask. (3) The race gap in employer callbacks is reduced when applicants have the chance to signal not having a record by answering “no,” which is consistent with theories of statistical discrimination. Policy Implications We conclude with a call to develop standards and best practices regarding inquiries about juvenile offenses, low‐level misdemeanor and traffic offenses, and the applicable time span. The need for such standards is made more apparent by the unevenness of criminal record questions across employees, establishments, and neighborhoods. We also suggest best practices for Ban the Box implementation to help combat potential statistical discrimination against African American men without records. Have you been convicted of a felony using your current name or any other name? If you do not answer this question, your application will not be considered. —Job application for laborer position at waste management company
Prior research has established the public recognizes the obstacles former prisoners’ face upon release from prison, and they support various offender reentry services. Still, we do not have a solid understanding of what affects public opinion on prisoner reentry nor is it clear that support for policies also extends to individuals’ willingness to engage with ex-offenders on a personal level. Drawing on a random sample of South Carolina residents, the current study examines the extent to which citizens’ belief in offenders’ redeemability influences their support for reentry initiatives at a policy and personal level. The results of this study show support for several types of reentry services, reveal that policy-level opinions and personal-level acceptance depend on the type of services, and demonstrate that belief in the redeemability of offenders is an important predictor of the degree to which the public embraces reentry programming.
Objectives Redemption scholars estimate that after an average of 7-10 years pass without a new arrest or conviction, a person’s criminal record essentially loses its predictive value. This article provides the first labor market and recidivism estimates of implementing a criminal background check decision guideline based on this redemption research. Methods The sample consists of provisionally hired job applicants in New York State’s healthcare industry with at least one prior conviction. A “10 years since last conviction” guideline situated within a highly formalized criminal background check process plausibly creates conditional random variation in clearance decisions, which allows for a regression model to estimate causal effects. Results Individuals cleared to work because of the 10-year guideline experience meaningful improvements in employment and earnings, but not recidivism on average. However, men do experience reductions in subsequent arrests, which appears to be driven by more complex factors beyond simply time since last arrest. Conclusions For some individuals, receiving clearance to work even a decade after their last conviction can have not only labor market benefits, but also important recidivism implications. Future research should explore the employment opportunity/recidivism trade-off in adjusting guideline threshold values and consider alternative redemption strategies.