Summary

6-Profilingemployeesonline--shiftingpublicprivateboundariesinorganisationallife.pdf

Home >English homework help >Article writing homework help >Summary

Profiling employees online: shifting public–private

boundaries in organisational life

Paula McDonald, Queensland University of Technology Paul Thompson, Stirling University, Queensland University of Technology Peter O’Connor, Queensland University of Technology Human Resource Management Journal, Vol 26, no 4, 2016, pages 541–556

Profiling involves the collection and use of online information about prospective and current employees to evaluate their fitness for and in the job. Workplace and legal studies suggest an expanded use of profiling and significant legal/professional implications for HRM practitioners, yet scant attention has been afforded to the boundaries of such practices. In this study, profiling is framed as a terrain on which employees and employers assert asymmetrical interests. Using survey data from large samples in Australia and the UK, the study investigates the prevalence and outcomes of profiling; the extent to which employees assert a right to privacy versus employer rights to engage in profiling; the extent to which organisations codify profiling practices; and employee responses in protecting online information. The findings contribute to a small and emerging body of evidence addressing how social media conduct at work is reconstituting and reshaping the boundaries between public and private spheres. Contact: Professor Paula McDonald, Queensland University of Technology, PO Box 2434, Brisbane, Qld 4001, Australia. Email: [email protected] Keywords: profiling; public–private boundaries; social media at work; employee privacy

INTRODUCTION

T he peer reviewed literature and popular media have reported the increasing use of ‘profiling’ by employers and HRM practitioners. Profiling, as defined in this article, refers to the collection of online information, often via social networking sites or generic

search engines, for the purpose of evaluating prospective employees and monitoring current employees with regards to their fitness for and in the job. Information gathered through profiling which is of potential interest to employers includes inappropriate comments or text, membership of certain groups and networks, communication skills, education, work history, professional affiliations, interests and lifestyle choices (Kluemper, 2013; Whitehall, 2012). Access to such information from employees’ online personas considerably extends traditional forms of evidence derived from reference checks and criminal background searches. This is because social exchange in an online environment which, although similar to traditional offline communication in that social interactions take place and information is exchanged, involves conversations which are preserved and subsequently accessible by others, including employers (Clark and Roberts, 2010).

Profiling has significant legal, ethical and professional implications for HRM practitioners (Davison et al., 2012), yet there has been relatively little discussion in the HRM literature itself. Rather, much of what we know about profiling is derived from law and organisational psychology/behaviour research and surveys of hiring professionals by consultancy firms. Studies from these fields point to an expanded use of online background searches in the recruitment process, and the use of such information to retain or disqualify applicants. However, they say little about the extent to which employees are aware of profiling practices and their outcomes; how employees assert their own as well as employer rights around its

HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016 541

Please cite this article in press as: McDonald, P., Thompson, P. and O’Connor, P. (2016) ‘Profiling employees online: shifting public–private boundaries in

organisational life’. Human Resource Management Journal 26: 4, 541–556

doi: 10.1111/1748-8583.12121

bs_bs_banner

use or the extent to which the boundaries of this type of social media conduct are transparent in organisational policy.

This article addresses such questions through a survey administered to two large samples from Australia and the UK. The study builds on McDonald and Thompson’s (2016) conceptualisation of profiling (and other social media conduct) as involving competing or asymmetrical terrains on which employers and employees make claims or rationalisations. As is evident in the following review, studies of profiling have generally defined the practice narrowly as one initiated by management during the recruitment process. Here, we expand the concept to include access to and utilisation of employee-owned profiles constructed through online personas (Ellerbrok, 2010) at any stage of the employment relationship. The study contributes to the small, emerging body of research which addresses how developments in social media use, and profiling in particular, are shaping public/private boundaries. We also consider aspects of codification, which is the process by which profiling and related employer- initiated social media practices are formalised in policies that can engender legitimacy and mutual obligation.

THE PRACTICE OF PROFILING

In general, online communications are not afforded the same protections as more traditional communications. The proliferation of social media technologies and associated behaviours has been and continues to be very rapid, with formal codification, community consensus and the law struggling to keep pace. Although the scale and scope of protection vary by country (see Finkin, 2002 for an overview), job applicants are often not covered by legislation relevant to the employment contract because the relationship with the employer occurs before a formal arrangement has been entered into.

Employees and prospective employees are, however, covered by discrimination law in many countries. Discrimination may result from profiling if an employer accesses and subsequently uses online information which relates to a protected ground to exclude an applicant or disadvantage an employee. Such information might include publicly available photographs which reveal physical biographical information such as sex or race, or non- physical information such as religion, sexual orientation or political persuasion. Research suggests that most individuals post accurate biographical information on their social media sites, much of which would be covered as protected grounds (Grasmuck et al., 2009). However, only very rarely are discrimination claims brought during or following a recruitment process, given the difficulties for a prospective employee to make an effective claim (Broughton et al., 2010) if the profiling was covert or undeclared by the employer.

Workplace studies suggest that searching for background information on job applicants as part of the hiring process is an increasingly common phenomenon. Somewhere between 12% and 50% of employers and/or HRM professionals surveyed across studies indicated they searched social networking sites for information about prospective employees (e.g. Brandenburg, 2008; Read, 2007). Of those who indicate they routinely use profiling, between a quarter and three quarters report that they use the information to disqualify applicants from further consideration (e.g. Brandenburg, 2008; Grasz, 2009).

The reasons reported for screening out applicants include posting provocative or inappropriate photographs, displaying poor communication skills, conveying information associated with alcohol or illegal drug use, revealing information that falsifies qualifications or credentials listed in a resume, posting disparaging or confidential content about former employers or work associates and concerns about associations with certain groups (Grasz,

Profiling employees online

542 HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016

2009; Mooney, 2010; Whitehall, 2012). Conversely, research has suggested that when the profiles of job applicants indicate a good fit between the individual and the company, the information improved their chances of being hired (Grasz, 2009). Hence, there may be mutual gains where carefully crafted, idealised identities offer tangible rewards for employees in the job market (Ellison et al., 2007).

To date, the emphasis has been on hiring, but the steady stream of cases concerning dismissals or discipline of employees for social media transgressions indicates a wider frame of managerial access to employee on-line data (see Bridges, 2015; Scutt, 2013). This kind of ad-hoc profiling is facilitated by the complex web of ‘friending’ and ‘following’ on Facebook and Twitter that traverses work and non-work networks. In either instance, the possibility of tension and conflict is present where employers access the private online personas of employees. This is especially the case when it is considered that the extent of disclosure associated with employees’ socially oriented profiles (such as Facebook), which they use to ‘create, exchange and disseminate information and ideas’, is often greater than their instrumental, career-oriented profiles such as LinkedIn (Ellerbrok, 2010: 202).

Legitimate practice or over-stepping the public–private boundary?

Employers and HRM practitioners put forward a range of interest-based assertions to rationalise the use of profiling. They argue that accessing such information protects them and their shareholders from negligent hiring (Blackwell, 2004), facilitates the acquisition of information about moral constitution and personality traits that may affect job duties (Morehead Dworkin, 1990), ensures employees have a clean online presence not likely to harm the organisation (Kluemper, 2013) and verifies information provided on the resume or application form (Brandenburg, 2008).

Although the online environment provides a promising source of applicant information on predictor constructs of interest, it is also fraught with potential limitations and challenges relevant to HRM (Brown and Vaughn, 2011). These limitations are both ethical, including violations of privacy, and legal, such as difficulties acquiring and using valid information which reliably predicts job performance and a lack of standardisation of information across applicants’ social media sites.

A prominent objection to profiling from employees is that the practice threatens the right to a private identity that is and should remain outside the purview of employers (McDonald and Thompson, 2016). However, the notion of privacy is destabilised in cyberspace because there are no physical boundaries that delineate behaviour and propriety (Levin and Sanchez Abril, 2009). Indeed, in the legal sphere, the entitlements of employees and job applicants to keep their personal online information concealed from employers is typically weighed against the rights of employers to monitor employees in order to reduce risks associated with legal liability, reputational damage, or reduced productivity. Whilst limitations have been set down in some jurisdictions, generally, few protections are afforded to employees or prospective employees who reveal their digital personas online on the basis that they are ‘publishers in a public realm’ (Howard, 2013: 1).

The use of profiling has also been questioned on the basis of the validity of the information gathered. It is feasible that information about drug use, discriminatory comments or misrepresentation of qualifications may accurately identify individuals who will demonstrate low levels of job performance or other negative organisational outcomes (Kluemper, 2013). However, despite safeguards embedded into many social networking sites, it is possible for checks to be inaccurate, mixing job applicants with the same name (Reicher, 2013), or using

Paula McDonald, Paul Thompson and Peter O’Connor

HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016 543

unfavourable or inaccurate information on sites without the target person’s knowledge or consent (Smith and Kidder, 2010). Bias can also occur whereby recruiters select applicants who are similar to themselves or who share common interests (Smith and Kidder, 2010). Compounding this problem is that the level of publicly available data obtainable by employers is often highly unstandardised. This is because some job applicants or employees may not use social media at all, whilst others, even if prolific users, may unevenly customise the degree to which information is made public to unintended audiences (Brown and Vaughn, 2011; Slovensky and Ross, 2012). As van Dijck (2013: 213) notes, ‘social media are not neutral stages of self-performance – they are the very tools for shaping identities’.

Concerns have also been raised around the transparency of profiling. With few exceptions (see Reicher, 2013), employers are not legally obligated to disclose to job candidates the sources of, or processes by which, they obtain information through profiling, nor how it affected subsequent outcomes (Carrington Davis, 2007). There has been little quantification of the extent to which employers communicate the boundaries of their profiling practices, such as through organisational policy. However, one study undertaken by a HR group in the US found that more than half of organisations have no formal policy with regard to social networking screening, whilst one in five had a (formal or informal) policy allowing the practice and one in four had a policy prohibiting the practice (Society for Human Resource Management, 2011). This is in contrast to evidence suggesting that the codification of employee-initiated social media and online conduct (e.g. critical online comments; private use of social media in work time) is becoming more common and more expansive, especially in large firms (Thornthwaite, 2013). In summary, there is emerging evidence that employers increasingly prohibit particular uses of social media by employees and use alleged infractions of policies to found misconduct allegations (Thornthwaite, 2013). However, there is scant evidence of the extent to which organisations codify employer-initiated social media practices such as profiling, nor which organisations are more or less likely to do so.

THE CURRENT STUDY

The practice of profiling has far outstripped research in the field (Kluemper, 2013). Although employer and HRM-targeted surveys have yielded important insights into what information is typically sought and utilised in decision making, the perspective of employees has been relatively neglected. Further, whilst rapid changes are evident in data retrieval technologies, knowledge of what employees, employers and the community more broadly consider acceptable is nascent.

Where research has examined responses to employer monitoring of personal online information, the focus has often been on so-called millennials. Several studies have indicated that younger employees are less sensitive to privacy concerns in the online environment than older employees, believing their communications are safe (e.g. Epstein, 2008), or that they are not willing to sacrifice internet participation to segregate their multiple life performances (Sanchez Abril et al., 2012). Hence, stated expectations of privacy appear to be somewhat inconsistent: employees generally want privacy from unintended employer eyes, and yet they share a significant amount of personal information online, knowing it could become available to employers and others (Sanchez Abril et al., 2012).

In this study, we investigate employer profiling across two samples (N=2000), via a large scale survey of working-age adults in the UK and Australia. The focal research questions were developed from neglected areas of extant work identified in the above review and build on dimensions of the profiling component of McDonald and Thompson’s (2016) broader model

Profiling employees online

544 HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016

of social media conduct at work, which conceptualises different forms of social media conduct as a series of contested terrains (see Figure 1). Relevant to the current study, profiling is rationalised by employers as a legitimate way in which online information can be used to facilitate the hiring of employees who exhibit ideal performativity and whose goals align with the organisation. Relatedly, the performance of connectedness in social networks is increasingly seen as a facet of employability in some occupations and companies (Gregg, 2009). In contrast, employees often claim that profiling threatens their interests in a private identity they claim should remain beyond employer scrutiny (McDonald and Thompson, 2016).

Building on this conceptual work and the broader literature on profiling, the survey addressed employees’ awareness of profiling, its outcomes, the extent of codification of the practice, how employees assert their own as well as employer rights in relation to the practice and the extent to which they protect their online information from current and future employers. The survey was administered in both Australia and the UK, countries with shared linguistic and cultural contexts but somewhat different industrial relations systems in that Australia is often considered to be more highly regulated. Explicit comparisons of patterns of survey responses across national contexts allows for an examination of which phenomena play out differently across different contexts and which may be more universal. We turn now to the empirical analysis.

METHODS

The study addressed the following research questions.

1. To what extent is the employer practice of profiling codified in organisational policy? What types of organisations do and don’t have policies?

2. From the vantage point of employees, what is the extent of profiling and what outcomes result?

3. To what extent do employees assert employee versus employers’ rights around the use of profiling?

FIGURE 1 Contested terrain of profiling as a dimension of social media in employment

Paula McDonald, Paul Thompson and Peter O’Connor

HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016 545

4. How sensitive are employees to privacy concerns in protecting their online information from current or future employers?

Sample and procedure

A 77-item survey was designed to explore employee behaviours and attitudes and organisational codification related to profiling, the posting of critical information online and private use of social media at work. Only data relevant to the former theme are presented in this article. Behavioural questions addressed the extent to which employees had experienced and witnessed profiling and what outcomes (hired, not hired) occurred and the degree to which employees use privacy settings to protect their online information from employer surveillance. Attitudinal questions addressed the extent to which employers had a right to search for information about prospective or current employees and the extent to which employees have a right to a private online identity. The codification of profiling activities was addressed by asking respondents whether or not their organisation had a policy.

The survey was designed and piloted by the authors and administered by an external research company with operations in Australia and the UK. The company had access to large panels (e.g. 300,000 in Australia) of adults 18years of age or over, who are registered to participate in a range of market and academic research surveys and receive incentives for doing so. Rules are in place to limit how frequently panel respondents can complete surveys. Methodological comparisons of panel surveys and telephone surveys show that panels can produce more reliable and consistent data estimates (Braunsberger et al., 2007). In both the UK and Australia, the survey was sent to a small sub-set of the total panel (around 2,000 individuals in each country) on a quota sample basis that ensured the final respondent pool would be broadly representative of the working age population and to keep weighting factors low in key groups such as males, younger individuals, and respondents living in regional areas, who tend to respond at a lower and slower rate than older, urban and female respondents. That is, potential respondents were matched to the actual age–gender–location profile of the adult working population in each country. Survey invitations were staggered across the time of day and day of week in order to further maximise representativeness.

The sample comprised 53% males and 47% females, with a mean age of 42 (SD=12.51). Thirty-six per cent of participants were aged between 17 and 34; 44% were aged between 35 and 54 and 20% were aged above 55. Seventy-three per cent were in permanent or ongoing work; 10% were in casual work; 11% were self-employed; 3% were on a fixed term contract and the remaining 5% were either agency workers, apprentices/trainees or volunteers. Fifty- three per cent of participants had supervisor/managerial responsibilities. Fifty-one per cent worked in organisations where nearly all staff used computers and another 12% where most did so. However, 17% indicated that computers were used by fewer than 20% of staff and 20% reported that computers were used by between 20% and 60%. Regarding organisation size, 48% of participants worked in large organisations (200+ employees), 25% in medium organisations (20–199 employees) and 27% in small organisations (1–19 employees). Details about industry and job type are summarised in Tables 1 and 2. We also compared the industry and job category profile of the sample with data from the Australian Bureau of Statistics (2015) and [UK] Office for National Statistics (2015) because along with age and gender which were controlled for in the administration of the survey, industry and occupation were key demographic variables of interest in revealing potentially divergent findings across different types of employees. The sample adequately represented those in higher skilled, computer- intensive roles (e.g. managers, professionals); slightly under-represented individuals in lower-

Profiling employees online

546 HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016

skilled, manual roles (e.g. machine operators, labourers) and slightly over-represented those in lower skilled administrative roles (e.g. administration and secretarial). Industry was broadly representative across all categories except wholesale/retail trade which was slightly under- represented, and Information/communication and Other service activities, which were both slightly over-represented.

Statistical analyses

For the first research question, a chi-square analysis and binary logistic regression assessed the extent to which profiling is coded in organisational policy, and identified what types of organisations do/do not have such a policy. For the second question, chi-square analyses assessed perceptions of the extent that profiling is used and its outcomes. For the third question, factor analysis and ANOVA determined the extent to which employees assert their own and employers’ rights around the use of profiling. Finally, to address the fourth question, factor analysis and ANOVA investigated the extent to which employees are sensitive to protecting their online information, and whether a range of demographic variables predict this sensitivity.

TABLE 2 Percentage of employees from each industry

Industry Per cent

Construction 4.9 Agriculture, forestry and fishing 1.8 Mining and quarrying 1.2 Manufacturing 7.5 Electricity, gas, air conditioning supply, water and sewerage 1.9 Wholesale and retail trade, repair of vehicles 5.8 Transportation and storage 4.5 Accommodation and food service activities 3.9 Information and communication 7.1 Financial and insurance activities 5.3 Professional, scientific and technical activities 7.1 Administrative and support service activities 5.3 Public administration and defence, social security 4.4 Education 10.7 Human health and social work activities 10.5 Arts, entertainment and recreation 3.5 Other service activities 14.1

TABLE 1 Percentage of employees in each job type

Job category Per cent

Managers, Directors, Senior Officials (e.g. corporate manager; chief executive) 11.2 Professional Occupations (e.g. scientist; engineer; architect) 26.1 Associate Professional/Technical Occupations (e.g. lab technician; paramedic) 8.6 Administrative and Secretarial Occupations (e.g. book keeper; secretary) 21.2 Skilled Trades Occupations (e.g. farmer; groundsman; mechanic) 9.9 Caring, Leisure and other Service Occupations (e.g. teaching assistant; vet nurse) 5.4 Sales and Customer Service Occupations (e.g. sales assistant; call centre worker) 9.0 Process, Plant and Machine Operatives (e.g. plant operator; van driver) 2.9 Elementary Occupations (e.g. farm worker; postal worker; cleaner) 5.8

Paula McDonald, Paul Thompson and Peter O’Connor

HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016 547

Results

The codification of profiling The majority of participants (N= 1570, 78.5%) indicated that they had knowledge about whether their organisation had a policy about using social media for profiling, whereas just under one quarter (N=430, 21.5%) indicated that they did not have such knowledge. Using data from the former group (N= 1570), Table 3 summarises the frequency of respondents from Australia and the UK who reported that their organisation did/did not have a policy. A slight majority of participants reported that their organisation had a policy (55.5%). There was no significant difference between participants from Australia and the UK (Chi2(1) = 1.21, p = .286). By way of comparison, more respondents (N= 1631, 81.6%) were aware of whether or not their organisation had a policy on employee-initiated conduct, such as making

TABLE 3 Frequency of respondents who reported their organisation did/did not have a policy about employers accessing employees or job applicant’s social media sites

Australia UK Total Frequency Per cent Frequency Per cent Frequency Per cent

Organisation has policy 418 54.1 453 56.8 871 55.5 Organisation does not have policy 355 45.9 344 43.2 699 44.5

TABLE 4 Organisational variables predicting the existence of social media profiling policies (n = 1559; binary logistic regression analysis)

B Exp(B) 95% CI Lower Upper

Country UK or Australia �.08 .92 .73 1.16 Industry type Construction

Agriculture, forestry and fishing .17 1.19 .44 3.18 Mining and quarrying .61 1.84 .56 6.05 Manufacturing �.24 .79 .42 1.45 Electricity, gas, steam and air conditioning supply, water

sewerage, waste management .23 1.26 .49 3.23

Wholesale and retail trade, repair of vehicles �.02 .98 .49 1.96 Transportation and storage .23 1.25 .61 2.58 Accommodation and food service activities .00 1.00 .48 2.11 Information and communication .05 1.05 .55 2.03 Financial and insurance activities .41 1.51 .75 3.06 Professional, scientific and technical activities .23 1.26 .66 2.40 Administrative and support service activities .47 1.61 .78 3.30 Public administration and defence, compulsory social security .99* 2.69 1.14 6.36 Education .43 1.54 .83 2.87 Human health and social work activities .31 1.36 .73 2.50 Arts, entertainment and recreation �.01 .99 .46 2.17 Other �.03 .97 .55 1.73

Org. size Medium (20–199) .68** 1.98 1.50 2.60 Large (200+) 1.94** 6.97 5.22 9.30

Computer use Some (20–40%) �.830* 2.29 .27 .70 Many (40–60%) �1.07** 2.91 .22 .54 Most (60–80%) �1.21** 3.36 .19 .47 All, or nearly all (80–100%) �.95** 2.57 .27 .56

* p < .05.

** p < .001.

X 2 = 358.57, 23 df, p < .001.

Profiling employees online

548 HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016

disparaging online comments, and a higher proportion (59.2%) indicated their organisation had a policy on this conduct.

In order to determine which demographic and grouping variables predicted codification of profiling, a binary logistic regression was conducted using data only from participants with knowledge of their organisation’s policy. Industry type, size of organisation, computer use (yes/no) and country (Australia/UK) were used as predictors in this analysis. The presence/absence of a profiling policy was the dependent variable. The reference category for the dependent variable was ‘no policy’; therefore, higher odd’s ratios (Exp(B)), as reported in Table 2, represent greater likelihoods of having a policy.

There was an overall relationship between organisation-characteristics and the presence/absence of a social media profiling policy (Chi2 (23)=358.57, p< .001). The primary predictors in this analysis were organisation-size and use of computers. Specifically, larger organisations and those where computers were used at least 20% of the time were significantly more likely to have a policy (see Table 4).

Industry type was not a strong unique predictor of the presence/absence of a social media profiling policy. Although preliminary univariate analyses indicated there were some differences in policy use based on industry type, these differences were fully accounted for by the tendency for these organisations to be large and have high computer use. The one exception was public administration and defence, which was more likely to have a policy on social media and profiling, even taking into account organisation size and computer usage.

Employee’s awareness of the extent of profiling and its outcomes Table 5 summarises descriptive statistics from four survey questions used to investigate the extent and nature of profiling from the vantage point of employees. Chi2 tests summarised in this table indicate whether differences exist between UK and Australian participants in their responses to these questions.

Less than 10% of respondents reported they had been subject to profiling. However, around a third of the sample reported that they did not know whether they had been profiled. This

TABLE 5 Employee perspectives on the extent and outcomes of profiling

Question Response Australia United Kingdom

Chi 2

n % n % p

1. Has an employer ever used online information about you to influence a hiring decision?

Yes 69 6.9 86 8.6 No 545 54.5 571 57.1 I don’t know 386 38.6 343 34.3 .08

2. If an employer has ever used online information about you to influence a hiring decision, what was the outcome of the hiring process?a

Hired 71 20.9 98 28.2 Not hired 33 9.7 32 9.2 I don’t know 236 69.4 218 62.6 .17

3. Has a potential employer ever asked you to provide them with your username or password to your social media site(s)?

Yes 33 3.3 67 6.7 No 910 91.0 882 88.2 I don’t know 57 5.7 51 5.1 .002

4. Have you ever witnessed or heard about an employer who used online information about a job applicant to influence a hiring decision?

Yes 290 29.0 253 25.3 No 562 56.2 624 62.4 I don’t know 148 14.8 123 12.3 .02

5. If you have witnessed or heard about an employer who used online information about a job applicant, what was the outcome of the hiring process?a

Hired 58 11.6 92 20.0 Not hired 128 25.7 104 22.6 I don’t know 312 62.7 264 57.4 .001

aData from participants answering ‘not relevant’ was not included in this table.

Paula McDonald, Paul Thompson and Peter O’Connor

HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016 549

pattern was evident across both the UK and Australia. For the individuals who reported being profiled, around twice as many applicants reported being hired as opposed to not hired. Again, this pattern was consistent across the UK and Australia.

Approximately four times as many individuals had witnessed or heard about an employer using profiling compared to those who experienced it directly. Slightly more participants from the UK (29.2%) reported they had witnessed/heard about profiling than participants from Australia (25.3%). Although small, this difference was found to be significant: Chi2 = 8.07, p =.018. In contrast to participants who had been personally profiled, participants who had heard about or witnessed profiling of other applicants tended to indicate that the applicant was not hired. Additionally, participants from Australia were significantly more likely than participants from the UK to report applicant failure (i.e. not being hired), rather than applicant success (i.e. being hired), where they had witnessed profiling: Chi2 =15.575, p= .001.

Employee’s perspective of their own and their employer’s rights to engage in profiling Table 6 summarises employee responses from the four survey questions used to investigate employee’s perspectives of their own and their employer’s rights to engage in profiling. This table indicates that participants tended to agree with items reflecting the belief that employees and applicants have the right to privacy, whereas participants tended to have more mixed attitudes regarding whether employers have the right engage in profiling.

To further explore participants underlying attitudes regarding employee rights, an exploratory factor analysis (EFA) was conducted on the four survey items. Two factors with eigenvalues greater than 1 emerged; the first factor captured an underlying belief that employees and applicants have a right to online privacy (termed ‘privacy’), whereas the second factor captured an underlying belief that employers have a right to conduct profiling (termed ‘search’). These factors were weakly negatively correlated (r=�.36), indicating that, generally, employees who strongly believe they have a right to online privacy tend not to strongly believe that employers have the right to search online. There were no significant differences between Australia and the UK on these attitudinal variables.

Two ANOVAs were then conducted in order to determine which demographic variables predict individual variation on the two attitudinal factors. The demographic variables used as predictors included gender, age, organisational size, education, supervisory/managerial

TABLE 6 Employee attitudes regarding their own and employers’ rights around the use of profiling

Question Response Australia United Kingdom

n % n %

1. Job applicants have a right to a private, online identity that should not be accessed by their employers, regardless of privacy settings

Agree 600 60.0 607 60.7 Neutral 264 26.4 277 27.7 Disagree 236 23.6 116 11.6

2. Employees have a right to a private, online identity that should not be accessed by their employers, regardless of privacy settings

Agree 608 60.8 602 60.2 Neutral 263 26.3 282 28.2 Disagree 129 12.9 116 11.6

3. Employers have a right to use online background information about job applicants to influence h iring decisions

Agree 458 45.8 436 43.6 Neutral 251 25.1 284 28.4 Disagree 291 29.1 280 28.0

4. Employers have a right to search for online personal information about current employees

Agree 423 42.3 385 38.5 Neutral 254 25.4 263 26.3 Disagree 323 32.3 352 35.2

Note: Agree/Strongly Agree and Disagree/Strongly Disagree categories were collapsed in this table.

Profiling employees online

550 HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016

responsibilities, employment status and level of computer use. Only two variables were found to significantly predict privacy concerns (Privacy). These were education level; the more educated, the more concern for privacy F(6, 1993)= 2.28, p< .05, and organisation size; employees from larger organisations tended to have more concern for privacy F(2, 1997) = 5.37, p< .05.

Several demographic variables were found to predict individual differences on the second variable (Search). Specifically, individuals were more likely to hold the belief that employers have a right to search online for extra information when they were: male F(1,1998)= 95.66, p < 001; working in professional and managerial positions (F(8, 1991)=5.82, p < .001); educated F(6, 1993)=4.27, p < .001; supervisors/managers F (1, 1998)= 14.34, p < .001 and used computers in their organisation at a high level F(4,1995)= 2.97, p< .05.

Sensitivity of employees to privacy concerns in protecting their online information from current/future employers Table 7 summarises employee responses to the three items measuring the extent to which employees actively manage their online activities. Across both samples, the majority of employees indicated that they actively manage their online activities at least ‘sometimes’. However, they were slightly less likely to use security settings in social media in order to prevent their manager/employer accessing their profile.

In order to assess the overall sensitivity of employees in regards to protecting their online information, an EFA was then conducted on the three items from Table 7. A single factor was identified with an eigenvalue greater than 1.

The items loading on this factor were then summed, in order to create a variable representing this factor, and an ANOVA was conducted in order to determine which demographic variables predict sensitivity. Several demographic variables were found to predict individual differences in the tendency for employees to manage their online activities with their employer in mind. Specifically, individuals were more likely to actively manage their online activities when they were: female F(1,1998)=14.03, p < .001; younger (particularly those aged 17–34), F(2, 1997) = 12.46, p < .001; ongoing employees as opposed to temporary workers, F(7, 1992)= 2.91, p < .05; employed in professional and managerial positions, F(1, 1998)= 46.68, p < .001; employed in larger organisations, F(2, 1997)=16.09, p< .001; supervisors/managers F(1, 1998)=46.68, p < .001; highly educated, F (6, 1993)= 9.81, p< .001 and employed in organisations that used computers F(4, 1995)= 28.67, p < .001. There were no significant differences between Australia and the UK on this variable t(1998)=1.50, p = .14.

TABLE 7 The extent to which employees actively manage their online activities

Question Response Australia United Kingdom

n % n % p

1. I manage my online and social media activities with my current employer in mind

Always or nearly always 429 42.9 372 37.2 .06 Often 166 16.6 195 19.5 Sometimes 141 14.1 148 14.8 Never or hardly ever 264 26.4 285 28.5

2. I manage my online and social media activities with future employers or employment in mind

Always or nearly always 355 35.5 276 27.6 .00 Often 154 15.4 182 18.2 Sometimes 159 15.9 157 15.7 Never or hardly ever 332 33.2 385 38.5

3. I use security settings in social media to try to prevent my manager or employer accessing my online content/profiles

Always or nearly always 290 29.0 301 30.1 .20 Often 120 12.0 135 13.5 Sometimes 104 10.4 122 12.2 Never or hardly ever 486 48.6 442 44.2

Paula McDonald, Paul Thompson and Peter O’Connor

HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016 551

DISCUSSION

Issues around profiling are indicative of the changing boundaries between public and private, work and non-work spheres. As Light notes, ‘A greater number of people are now engaging with SNSs, and for many, these activities are becoming entwined with their employment status’ (2014: 93). Yet profiling has proceeded largely be stealth. By this we do not simply mean that a good deal of managerial action occurs under the radar, but that accessing employee data for and in the job is running ahead of, or parallel to, policy and perception. This helps to frame some of the findings with respect to the reach and legitimacy of managerial practices and employee privacy rights discussed further below. Practice has also tended to run ahead of academic analysis. This study, which frames social media conduct as a terrain on which employees and employers assert asymmetrical concerns, contributes to this significant gap in the work and organisation literature, and in particular, knowledge of how social media conduct at work is reconstituting key aspects of the employment relationship.

Prevalence, awareness and responses to profiling

This study is one of the first attempts to investigate to what extent, and how, employees experience, witness and respond to profiling. Less than one in 10 employees indicated they had been profiled in a recruitment process and around a third did not know whether they had been subject to such surveillance. These findings suggest a marked contrast between employees’ awareness of whether they have been individually profiled, and the extent of actual profiling conduct suggested by surveys of HR managers and other recruitment specialists which indicate it is a majority practice (Brandenburg, 2008; Read, 2007). These discrepancies between what organisations do to gather information about prospective employees, and the extent to which employees are aware of such actions, are consistent with the largely covert nature of the practice (Carrington Davis, 2007), at least for the targets involved.

The survey also showed, however, that employees had a much greater awareness of applicant profiling as it was applied to others, with four times as many individuals indicating they had witnessed or heard about profiling than those stating they had experienced it personally. Relatedly, those who witnessed applicant profiling were much more likely to indicate that the applicant had not been hired than those experiencing this form of profiling directly. A possible explanation of this finding is that those who are not hired in a recruitment process may not know they were profiled because they are outside the organisational environment, and hence have little knowledge of recruitment data gathering and decision- making processes. In contrast, those who witness profiling may be more likely to do so from within an organisation and therefore have greater awareness of the outcome either way. The greater likelihood of the applicant not being hired in cases where profiling was witnessed aligns with previous research suggesting profiling is frequently used to screen out applicants deemed to be unsuitable (e.g. Brandenburg, 2008). This trend was amplified in Australia compared to the UK, although this was virtually the only finding where between-country differences were statistically significant. The very similar results found across these two national contexts, coupled with the striking parallels seen in employment legal disputes in different countries, suggest the dimensions of contestation around social media conduct in the workplace may, to some degree, traverse national, regulatory and cultural boundaries.

The study also sought to determine what employees believed were the boundaries of profiling in terms of the limits on employer monitoring and surveillance. Respondents asserted a stronger overall concern for their right to privacy in the online environment than the right of

Profiling employees online

552 HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016

employers to search for online information about current or prospective employees. Although this was the case on average across all employees, the finding was stronger amongst highly educated men who worked in professional and managerial roles. However, individuals with higher educational levels and who worked in professional/managerial roles in larger organisations, were also more likely to manage their private online information in a way that protected it from current and prospective employers. Together, these results suggest that those with higher awareness of and familiarity with social media were more likely to actively manage their private information, despite acknowledging that employers were entitled to search for such information. This finding may be explained by a greater familiarity with the technologies themselves, including how to manage privacy settings, and/or the greater salience of public– private boundaries in professional and managerial roles, which are increasingly characterised by blurred employment and personal spheres, or what Vitak et al. (2012) refer to as ‘context collapse’.

An unexpected finding in terms of the extent to which employees manage their online information was that younger individuals do this more actively than older employees. This challenges some research which indicates so-called millennials are less sensitive to privacy concerns (e.g. Epstein, 2008) and have higher expectations of network privacy or audience segregation (Sanchez Abril et al., 2012). The notion of audience segregation is a useful one in explaining why profiling is so contentious. Irrespective of demographic characteristics, the maintenance of contextual boundaries is often desired by employees to separate their professional and person lives (Pike et al., 2013). Yet profiling may breach expectations of audience distinction, threatening employees’ interests in having a private identity beyond employer scrutiny.

Codification of profiling and policy implications

Larger organisations, and organisations whose functions involved greater computer use, were more likely to have developed policies determining the acceptable nature of profiling. Public and private sector organisations will generally have larger and more active HR and public relations functions with experience and capacity of policy activism in other code of conduct spheres (McDonald and Thompson, 2016). Organisations in the industry category public administration/defence were especially likely to have developed policies. The more proactive approach in these sectors in formalising what they see as appropriate social media conduct is likely to reflect a greater awareness of risk and closer proximity to public policy discourses.

Moving to the employee perspective, around three-quarters of the sample either did not know whether their organisation had a profiling policy in place or indicated that it did not have a policy. There is, seemingly, an imbalance between the transparency and effective communication of policies around profiling, compared to the sometimes aggressive pursuit of codes determining employee-initiated social media behaviours. Scant empirical attention has been afforded to the content and reach of employer policies directing social media conduct and this warrants further attention. However, the evidence available suggests that policies are edging towards a greater regulation of employees’ private lives, intentionally or unintentionally limiting their freedom of expression, especially to explore issues of mutual concern among colleagues during off-duty conversations (Thornthwaite, 2013).

Uneven practices and perceptions with respect to codification, profiling and privacy suggest a number of implications for HR policy related to transparency, legitimacy, ‘friending’ practices of managers, confidentiality and safe storage, and more generally a risk–benefit analysis of the scope and content of screening (Davison et al., 2012; Kaupins and Park, 2011; Slovensky and

Paula McDonald, Paul Thompson and Peter O’Connor

HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016 553

Ross, 2012). There is also a need for a wider workplace conversation about the relevance and reach of profiling. For example, Charlesworth (2003) asserts that the intrusion of profiling practices should be offset by a greater utility to the employer or society, use the least intrusive measures possible to achieve the desired outcome and apply the measure equally to similarly situated job applicants. More generally, if new practices are to be sustainable in a context of changing public/private boundaries where there are greater expectations of expanded social media use in employment, transparency and parsimony should also be accompanied by enhanced reciprocity and a better balance between employer requirements and ‘safe spaces’ for employees. For these reasons, conversations should also include employees, their unions and professional bodies. There is some evidence from our results and other sources (e.g. Light, 2014) that some employees are becoming more aware and selective about who and what they connect to and talk about, but the picture is very uneven considering the risks to job security and privacy.

CONCLUSIONS AND FUTURE RESEARCH

Recent developments in the field of social media conduct at work have been patchy and arbitrary, with recurrent legal disputes reinforcing the sense of a workplace politics of time and place that are increasingly contentious and contested (McDonald and Thompson, 2016). The findings reported here have ventured into new empirical territory, revealing the prevalence and codification of profiling, the extent to which employees perceive the practice as fair, and how they consequently respond. In adopting an explicitly employee-focused perspective, the study makes a considerable contribution to the existing literature on profiling. Extant work has focused primarily on the prevalence and nature of the practice as reported by HR practitioners and recruitment specialists, and/or on concerns about profiling, especially around validity, transparency and consistency.

Continuing to assess the scope and extent of profiling activities on the employer terrain is important, because employers’ interests in recruiting engaged employees who exhibit ideal performativity are unlikely to diminish. Future research which examines monitoring activities should also include a focus on the technologies associated with and approaches to tracking social media, if and how profiling is supervised in organisational environments and the kinds of occupational roles that are targeted. However, there are important implications for HR personnel in not only understanding how profiling is practised, but also how prospective and current employees perceive and respond to having their personal information monitored. This is especially the case because, as the data here show, such monitoring is frequently not transparent to or is purposefully concealed from employees.

Acknowledged limitations of web surveys include that respondents may falsify their demographic information and that they suffer from coverage error which is the difference between the defined target population of interest and the population frame used for the study (Couper, 2000). Mitigating these weaknesses was that the panel used for this research was specifically recruited for online research and quota sample techniques were adopted (Braunsberger et al., 2007). What this or other survey evidence cannot tell us however, is how profiling is interpreted, enacted and responded to by managerial and institutional actors who either have responsibility for monitoring and regulating social media boundaries and behaviours or who are subject to them. Survey research is also limited in revealing the kinds of information which are considered problematic by employees or employers, and how actors weigh up the limits of the others’ interests, that is, in either performativity or private identity.

Profiling employees online

554 HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016

Future research addressing these issues would help inform policy and regulation on opportunities and limits to voice and surveillance across industry/organisational contexts, which are in turn conditioned by available power resources. Although most employees to some extent accept employers’ entitlements to search for personal online information, emergent rights around profiling are essentially one sided, in that the practice is opaque and subject to relatively few codified boundaries, despite the myriad of limitations detailed in the literature. In contrast, what is urgently needed is a rights agenda characterised by greater transparency and reciprocity which will help build community consensus and a greater balance of interests between employees and employers.

The research was funded by the Australian Research Council, FT120100635. There is no interest to disclose.

REFERENCES

Australian Bureau of Statistics (2015). ABS Labour Force, Australia. Catalogue Number 6291.0.55.003. Blackwell, C. (2004). ‘Current employee privacy issues’. The Journal of Applied Management and

Entrepreneurship, 9: 1, 113–118. Brandenburg, C. (2008). ‘The newest way to screen job applicants: a social networker’s nightmare’. Federal

Communications Law Journal, 60: 3, 597–626. Braunsberger, K., Wybenga, H. and Gates, R. (2007). ‘A comparison of reliability between telephone and

web-based survey’. Journal of Business Research, 60: 7, 758–764. Bridges, C. (2015). Social media and unfair dismissal: bad news for employees. Available at http://www.

keepcalmtalklaw.co.uk/social-media-unfair-dismissal-bad-news-for-employees/ (accessed 1 July 2016). Broughton, A., Higgins, T., Hicks, B. and Cox, A. (2010). Workplaces and Social Networking. The Implications

for Employment Relations, Brighton: Institute for Employment Studies. Brown, V. and Vaughn, E. (2011). ‘The writing on the (Facebook) wall: the use of social networking sites in

hiring decisions’. Journal of Business Psychology, 26: 2, 219–225. Carrington Davis, D. (2007). ‘My space isn’t your space: expanding the Fair Credit Reporting Act to ensure

accountability and fairness in employer searches of online social networking services’. Kansas Journal of Law and Public Policy, 16, 237–256.

Charlesworth, A. (2003). ‘Opinion: privacy, personal information and employment’. Surveillance and Society, 1: 2, 217–222.

Clark, L. and Roberts, S. (2010). ‘Employer’s use of social networking sites: a socially irresponsible practice’. Journal of Business Ethics, 95: 4, 507–525.

Couper, M. (2000). ‘Web surveys: a review of issues and approaches’. Public Opinion Quarterly, 64: 4, 464–494.

Davison, H., Maraist, C., Hamilton, R. and Bing, M. (2012). ‘To screen or not to screen? Using the internet for selection decision’. Employee Responsibility and Rights Journal, 24: 1, 1–21.

Ellerbrok, A. (2010). ‘Empowerment: analysing technologies of multiple variable visibility’. Surveillance and Society, 8: 2, 200–220.

Ellison, N., Steinfield, C. and Lampe, C. (2007). ‘The benefits of Facebook “Friends”: exploring the relationship between college students’ use of online social networks and social capital’. Journal of Computer-mediated Communication, 12: 4, 1143–1168.

Epstein, D. (2008). ‘Have I been Googled? Character and fitness in the age of Google, Facebook and YouTube’. The Georgetown Journal of Legal Ethics, 21: 1, 715–727.

Finkin, M. (2002). ‘Information technology and workers’ privacy: the United States law’. Comparative Labor Law and Policy Journal, 23: 4, 471–506.

Grasmuck, S., Martin, J. and Zhao, S. (2009). ‘Ethno-racial identity displays on Facebook’. Journal of Computer-Mediated Communication, 15: 1, 158–188.

Grasz, J. (2009). Forty-five percent of employers use social networking sites to research job candidates, careerbuilder survey finds. CareerBuilder Press Releases. Available at http://www.careerbuilder.com/

Paula McDonald, Paul Thompson and Peter O’Connor

HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016 555

http://www.keepcalmtalklaw.co.uk/social-media-unfair-dismissal-bad-news-for-employees/

http://www.careerbuilder.com/share/aboutus/pressreleasesdetail.aspx?id=pr519andsd=8%2f2009anded=12%2f31%2f2009andsiteid=cbprandsc_cmp1=cb_pr519

share/aboutus/pressreleasesdetail.aspx?id=pr519andsd=8%2f2009anded=12%2f31%2f2009andsiteid= cbprandsc_cmp1=cb_pr519 (accessed 10 April 2015).

Gregg, M. (2009). ‘Banal bohemia blogging from the ivory tower hot-desk’. Convergence: The International Journal of Research into New Media Technologies, 15: 4, 470–483.

Howard, L. (2013). Speech to the Social Media Disasters Roundtable 28 February 2013, Melbourne: Law Institute of Victoria Legal Symposium.

Kaupins, G. and Park, S. (2011). ‘Legal and ethical challenges of corporate social networking’. Journal of Communications Research, 2: 2/3, 119–145.

Kluemper, D. (2013). ‘Social network screening: pitfalls, possibilities, and parallels in employment selection’, in T. Bondarouk and M.R. Olivas-Luján (eds), Social Media in Human Resources Management (Advanced Series in Management, Volume 12), Bingley: Emerald Group, 1–21.

Levin, A. and Sanchez Abril, P. (2009). ‘Two notions of privacy online’. Vand Journal of Entertainment and Technology Law, 11: 4, 1001–1008.

Light, B. (2014). Disconnecting with Social Network Sites, Basingstoke: Palgrave Macmillan. McDonald, P. and Thompson, P. (2016). ‘Social media(tion) and the reshaping of public/private boundaries

in employment relations’. International Journal of Management Reviews, 18: 1, 69–84. Mooney, D. (2010). ‘Employer on the web wire: balancing the legal pros and cons of online employee

screening’. Idaho Law Review, 46: 3, 733. Morehead Dworkin, T. (1990). ‘Protecting private employees from enhanced monitoring: legislative

approaches’. American Business Law Journal, 28: 1, 59–85. Office for National Statistics (2015). Labour Market Statistics, September 2015. EMP04: Employment by

Occupation. Available at http://ons.gov.uk/ons/publications/re-reference-tables.html?edition=tcm% 3A77-375224#tab-Employment-tables.

Pike, J., Bateman, P. and Butler, B. (2013). ‘Dialectic tensions of information quality: social networking sites and hiring’. Journal of Computer-Mediated Communication, 19: 1, 56–77.

Read, B. (2007). ‘Online’. Chronicle of Higher Education, 53: 19, A31. Reicher, A. (2013). ‘The background of our being: internet background checks in the hiring process’.

Berkeley Technology Law Journal, 28: 1, 115–153. Sanchez Abril, P., Levin, A. and Del Riego, A. (2012). ‘Blurred boundaries: social media privacy and the

twenty-first century employee’. American Business Law Journal, 49: 1, 63–124. Scutt, M. (2013). ‘Misuse of social media by employees’. Available at http://www.infolaw.co.uk/newsletter/

2013/09/misuse-of-social-media-by-employees/ (accessed 1 July 2016). Slovensky, R. and Ross, W. (2012). ‘Should human resource managers use social media to screen job

applicants? Managerial and legal issues in the USA’. Info, 14: 1, 55–69. Smith, W. and Kidder, D. (2010). ‘You’ve been tagged! (Then again, maybe not): employers and Facebook’.

Business Horizons, 53: 5, 491–499. Society for Human Resource Management (2011). SHRM survey findings: the use of social networking

websites and online search engines in screening job candidates. Available at http://www.shrm. org/Research/SurveyFindings/Articles/Documents/PART%202_Social%20Media%20Flier_FINAL.pdf (accessed 26 April 2015).

Thornthwaite, L. (2013). ‘Chilling times: labour law and the regulation of social media policies’. Labour Law Research Network Inaugural Conference, Barcelona, June.

Van Dijck, J. (2013). ‘“You have one identity”: performing the self on Facebook and LinkedIn’. Media, Culture and Society, 35: 2, 199–215.

Vitak, J., Lampe, C., Gray, R. and Ellison, N. (2012). ‘Why won’t you be my Facebook friend? Strategies for managing context collapse in the workplace’. Proceedings of the 7th Annual iConference, Toronto, Canada, 555–557.

Whitehall, M. (2012). ‘Better safe than subjective: the problematic intersection of prehire social networking checks and title VII employment discrimination’. Temple Law Review, 85: 1, 229–267.

Profiling employees online

556 HUMAN RESOURCE MANAGEMENT JOURNAL, VOL 26, NO 4, 2016

http://www.careerbuilder.com/share/aboutus/pressreleasesdetail.aspx?id=pr519andsd=8%2f2009anded=12%2f31%2f2009andsiteid=cbprandsc_cmp1=cb_pr519

http://ons.gov.uk/ons/publications/re-reference-tables.html?edition=tcm%3A77-375224#tab-Employment-tables

http://www.infolaw.co.uk/newsletter/2013/09/misuse-of-social-media-by-employees/

http://www.shrm.org/Research/SurveyFindings/Articles/Documents/PART%202_Social%20Media%20Flier_FINAL.pdf