Chapter 1 A bad day at the office
1 We are very grateful to Emma Calvert who introduced us to this method of utilising the qualitative data gathered as part of a large social survey (see O’Connell et al. 2007).
2 The roots of concepts of workplace bullying lie in a definition originally used to measure the construct in school children’s behaviour (Olweus 1991, cited in Nielsen et al. 2010). In Chapter 8, we show how employees in one of our case study organisations used an implicit contrast with the behaviour of the playground (‘we treat each other as adults’) to explain how trouble at work was minimised.
3 It is frequently found that dissatisfaction levels among unionised workers are higher than among their non-unionised counterparts (Bender and Sloane 1998). Higher reports of bullying amongst union members might reflect their higher expectations, their greater awareness of bullying as a public issue, and/or the fact the employees who face problems at work may be more likely to join trade unions in order to get help to address them.
Part two
1 For example, see Notelaers et al. 2006. Such studies have found that there is poor overlap between the groups of ‘bullied’ workers when measured on each different approach. In particular, less than half of the respondents who are counted as victims of bullying by researchers analysing responses to the NAQ report themselves as bullied (Lutgen-Sandvik et al. 2007; Notelaers et al. 2006; Salin 2001).
2 Respondents for the survey were identified by screening participants in Taylor Nelson Sofres’s (TNS) face-to-face Omnibus survey. A representative sample of around 2,000 adults per week in Britain (England, Wales and Scotland) was interviewed by Omnibus. It was carried out using a quota sample, with sample points (and addresses within these sample points) selected by a random location methodology. TNS fieldworkers used the CAPI (Computer Assisted Personal Interviewing) method to administer the survey in respondents’ households.
3 The empirical story behind our decision is a little complicated as it relates to the 21 items having both ‘original’ versions and also ‘confirmed’ versions (the latter representing an additional opportunity for respondents to confirm whether their previous answers were indeed correct and consequently are more accurate). Therefore there were two factor analyses and the differences between them, along with the differences between the factor loadings within each factor analysis, suggested certain items could belong to more than one factor. Our decisions were made taking into account our understanding of extant research, our own empirical findings, and results from a factor analysis of Fair Treatment at Work Survey (FTWS) data.
Chapter 2 Fairness and rationality at work
1 Are these rogue results? We asked three of these questions in the FTWS (undertaken a year later and with random sampling instead of quota sampling): 12 per cent of employees/recent employees had been pressurised to work below their level of competence and 13 per cent complained of unmanageable workloads. This gives us confidence in the reliability of our findings. The greater variation between the two survey measurements of employers’ use of improper procedures (23 per cent in the BWBS versus 17 per cent in the FTWS) probably results from the differing content of the questionnaires used in each survey. The BWBS had no questions on knowledge or awareness of employment rights or on any other problems than ill-treatment. The FTWS had many more alternative questions where respondents might record their dissatisfaction with their employers’ procedures. For example, 8 per cent agreed with ‘your employer not following a set procedure when dealing with a grievance of other work-related problem you had’ (Chart 6.1 in Fevre et al. 2009). It is possible that perceived overlap may have reduced responses to the question on ill-treatment. In any event, what we find in both of these nationally representative samples with face-to-face interviewing is much lower incidence rates for these types of problems at work than are often quoted. For example, Hoel and Cooper (2000) reported in their study that nearly a third of employees reported ‘pressure from someone else to do work below your level of competence’ now and then or more frequently within the past six months. The comparison figure from the BWBS was 13 per cent (and 11 per cent from the 2008 FTWS).
2 The stress and bullying literature seems to assume job satisfaction or disengagement is caused by ill-treatment. We are sure this is not the whole story, just as we are sure that the bullying literature does not tell the whole story of the relationship between witnessing ill-treatment and suffering ill-health when it assumes that this is just a matter of effects on health.
3 Similar patterns were recorded in the FTWS.
4 See pp. 30–1 in the introduction to Section Two – these were the respondents who reported three or more types of ill-treatment and tended to have more intense exposure. For example, a third (87) of the 265 respondents who were asked questions about unmanageable workload said this happened monthly or more frequently. It can be recalled that we made our own judgements about what form of ill-treatment was most important, and so we tended to collect more evidence on violence and other forms of ill-treatment where the numbers involved were relatively small. Thus, there were less data about how unreasonable treatment affected the troubled minority compared with the sample as a whole. For example, we asked 265 respondents follow-up questions about unmanageable workload, but we also asked 689 respondents about incidents where people had shouted at them or lost their temper.
5 In the 27 EU countries there are roughly twice as many male as female managers (Eurostat News Release 32/2008 – 6 March 2008).
6 There may well be psychological explanations for the failure of individuals to recognise their own shortcomings, but the point remains that people expect reasonable treatment in the workplace even if they expect it nowhere else.
7 Note that when employees’ views and opinions were ignored, this was not linked to being excluded from a group or clique at work.
8 As Figure 7 suggests, there was little suggestion that troubled workplaces were more likely to be found in the public sector. Indeed, employees in education were less likely to experience, and witness, unreasonable treatment. Bivariate analysis for witnessing unreasonable treatment suggested that public sector workers of any kind were more likely to report it but only in two of the eight types of unreasonable treatment.
Chapter 4 Violence and injury at work
1 For example, only 0.9 per cent of working adults reported experience of physical violence during the past year in the 2002–3 British Crime Survey (Upson 2004).
2 For example, the FTWS found that 4 per cent experienced physical assault (Fevre et al. 2009), and our finding is also broadly comparable to the estimate of the 2005 European Working Conditions Survey (EWCS) that 3.6 per cent of the working population in the United Kingdom were subject to physical assault from people at work (presumably colleagues), and a further 7.3 per cent had experienced physical violence at work perpetrated by non-colleagues.
3 The cognitive testing interviews that were undertaken as part of the survey development work suggested that the question about physical violence was interpreted in a straightforward way as intentional interpersonal assaults, rather than structural violence (Jones et al. 2011).
4 There are many ways we could group the four categories of perpetrators; however, this is the only one that is accurate in so far as it reflects the actual differences between clients or customers and the other three types. It is important to not imbue these terms with assumptions about levels of intimacy, frequency of contact or power differentials between the type of perpetrator and respondent. For example, it is reasonable to suppose that some clients/customers would be very familiar to respondents, even inescapable in terms of contact if, for example, they lived in the care home where the respondent worked. However, some clients might feature prominently in the lives of respondents, and wield enormous power over them in their working lives, if, for example, they were long-term clients and the respondent was relatively new to the organisation. These examples only serve to illustrate how much we still do not know about the nature of the relationship between the respondent and perpetrator, despite our best efforts to collect data in this regard (and within the financial constraints of our project).
5 To avoid complication, we here include violence perpetrated by ‘employers’ under the broader category of ‘employee’ violence, in order to demonstrate the key distinction between violence carried out by people working in the same organisation (whether employees or the employer) and that carried out by those who do not work in the organisation.
6 Based on analysis where the perpetrator was a client in any of the three incidents.