Theoretical approaches to the study of bullying at work

Charlotte Rayner Staffordshire University Business School, Stafford, UK, and Michael Sheehan and Michelle Barker Griffith University, Brisbane, Australia
Keywords: Bullying, Methodology, Theory

Type of Article: Wholly theoretical

This article provides a brief overview and commentary on theoretical approaches discussed in the articles by Einarsen, Liefooghe and Olafsson, and Rayner in the theoretical section of this special issue. In highlighting the difficulty in defining bullying, this article suggests that definitional boundaries ought to be all encompassing and open for discussion as a way of further developing the debate about workplace bullying. A discussion of methodological approaches, including those used by the authors in the theoretical section then follows. The article suggests that a collaborative approach between employers and researchers would help further address the problem of workplace bullying.

Content Indicators: Readability**, Practice Implications*, Originality**, Research Implications***

The articles in this section provide review and comment on theoretical approaches to bullying at work. These papers reflect an increasing sophistication within the field, and point to an exciting potential for further research. Within them one can observe how the outputs of research in bullying at work give a clearer picture of the construct. Concurrently, one can observe fragmentation as writers discuss the complexity of the phenomenon. These themes will be developed within this article.

Bullying in the workplace is attracting more media attention (Sheehan, 1999); legal attention (Sheehan et al., 1999), and academic interest (of which this special issue is evidence). Patterns appear internationally which evidence an escalating profile of a new field of attention, similar to that of sexual harassment (Fitzgerald and Shullman, 1993) more than a decade ago.

As with sexual harassment and stress, there is no agreed definition of bullying. In reality, three application areas for definition appear to have emerged. First there are those that deal with people with severe trauma resulting from being bullied at work (e.g. Zapf et al., 1996). Second there are those which explore the concept within the general working population (e.g. Liefhooghe and Olafsson, 1999). Finally there are those which are used in incidence studies (e.g. Rayner, 1999).

Incidence studies, at some point, need to identify whether or not their respondents are being bullied. By using rules for "counting" the bullied, incidence studies inevitably generate a bi-modal construct. Definitions vary between countries and studies, with differences being well reviewed elsewhere (e.g. Hoel et al., 1999; Keashly, 1998). Concrete issues such as longevity and frequency of negative (bullying) behaviours are likely to be the first parameters that will gain international agreement. This would represent progress from our current situation of non-comparable incidence studies.

Which behaviours are included within a definition presents a considerably more complex problem. This applies to incidence studies, and also includes the application area of exploring the views of the wider (non-bullied) population. Liefooghe and Olafsson's article (1999) is a good example of such a generalised application area. Their argument that the behaviours that are perceived as "bullying" may differ from one organisational context to another is well made. While the application area has yet to be explored systematically, the notion is sensible. Such an argument would be exponentially stronger if it were applied to trying to establish an international set of behaviours for use in cross-national incidence studies.

The study of bullying at work has its roots in clinical work with severely traumatised people (Leymann, 1990) which presents a radically different context for the application of definition than that of the general public. In this context "referral for treatment" would appear to be an initial process within defining such a group. Thus the process of definition may rest more heavily on clinical professionals than on the bullied individuals or their workmates. Data collection from such groups continues to be useful (e.g. Vartia, 1996).

Bullying at work is not necessarily an all or nothing experience, as Einarsen (1999) correctly points out. In this way the definition constructs for incidence studies may not reflect real life. Liefooghe and Olafsson's article exposes the richness that may be gained when in a qualitative research environment, the definition boundaries are allowed to rest where appropriate. Their work with the general working population explores the way such individuals make sense of the general issue of bullying.

Einarsen (1999) also provides a wide-ranging review of literature that prompts questions regarding the "domain" (Liefooghe and Olafsson, 1999; Keashly, 1998), since a definition should be able to locate an individual within the domain the definition addresses. Our current definitional parameters are dominated by frequency; longevity; reaction to behaviours, and power differences. It is likely that these parameters do not go far enough to encompass some of the more process-oriented work now emerging (e.g. Neidl, 1996; Leymann, 1996; Einarsen, 1999).

This said, a definition needs to be fit for purpose, and perhaps one way forward is the requisite for clarity regarding what is the purpose of a particular definition. Others have commented (e.g. Einarsen, 1999) that we are using one word for a variety of situations, and that we need to broaden our vocabulary to reflect a wide range of uses and contexts (Hoel et al., 1999). It is questionable that the same word is applied to a severely traumatised target of bullying who has left their job and also to someone who is able to cope in some way and who remains in their job even though this is not a pleasant experience.

The debate on definition has been invigorated in recent years. We would suggest that this debate should be an open discussion (as evidenced by several papers in this issue) so that those in the field can avoid an early entrenchment in exclusive "camps" of particular opinion. The issue of language for the "naming" of the phenomenon (or phenomena) may be an issue to address sooner rather than later.

Central to the current issues in bullying research are those which revolve around methodology. As our perspective widens, so does our perception of what might need to be studied, and therefore the range of methodologies from which to choose also warrants consideration. Liefooghe and Olafsson's article (1999) has widened the scope of research to individuals who are not bullied. In the past others have reported on the non-bullied (e.g. Zapf et al., 1996; Rayner, 1997), but their purpose has been to provide benchmarks for considering the bullied who remained the focus. If we accept that bullying is behaviour which is accepted as a phenomenon in working culture (Archer, 1999) then it is imperative that we need to understand the processes of "making sense" of bullying within the working population at large. A broader range of methodologies beyond the interview and questionnaire can be expected and is starting to emerge, such as the use of focus groups (IRS, 1999), Critical Incident Technique (Liefooghe and Olafsson, 1999) and the use of vignettes (Keashly et al., 1994).

It is unlikely that incidence studies will deviate from their usual methodology of large-scale studies using postal questionnaires. Perhaps stronger collaboration between researchers may provide cross-cultural comparability across factors, ideally building on established constructs such as those generated by Hofstede (1980). Thus large-scale studies may be able to explore difference within a well-researched arena.

Within the more recent articles in bullying at work is an accent on process (Einarsen (1999) provides a useful review). Such an emphasis is crucial to our understanding of the phenomenon. The implications for methodologies point to qualitative study rather than quantitative study. Researchers often mention the attribution process (e.g. Rayner, 1999) as one that may confound accurate reporting, and indeed, attribution may be better assessed and tackled in a qualitative setting. The quest for "the truth" may be "out there" (especially in a legal context), but in our opinion, researchers in this field are realising that they must work with different realities and perceptions as part of the process that we are studying. As such the methodologies employed may be more flexible than if "the truth" were the only target.

Process models are usually transactional by nature. Some models (e.g. Leymann, 1996) identify the stages to which the target of bullying is subjected. A clear trend can be found in this issue towards including the target of bullying behaviour as part of a dynamic process, rather than as a passive recipient. Such a shift has major implications and its results will need careful management. It is difficult to escape the issue of blame in bullying at work. There is a danger that through assigning targets a positive role in a dynamic process, their actions are interpreted as being those of collusion or complicity. Such interpretations may take the focus away from abusive bullies and organisations that provide reward for mistreatment of others. Models such as those of transactional stress (Lazarus, 1976) may provide us with clues for how to tread the "process" path and enrich our work in coping strategies.

A further layer of complexity is noted, in that the articles in this issue reflect work at a number of different "levels". Lewis considers a wide societal focus. Others such as Archer consider the organisational perspective. All embrace the interpersonal nature of bullying. Einarsen raises this difference at an analytical level amongst the studies he reviews and correctly, in our view, sees such difference as a solution to some apparent conflicts in data. We would take his point further and suggest that methodologies that are able to link across levels and deepen our understanding of a process that genuinely spans from the interpersonal to the societal will greatly benefit the field. This argument implies the use of multiple measures, and calls for researchers who are willing to work in a multidisciplined way. There will always be room for specialists, but it is our opinion that the field of workplace bullying will see those researchers who can span across disciplines thrive.

Another key point relating to methodology is that of access. In order to study the bullying process, it is not ethically possible to simulate the situation in the laboratory!Worldwide, researchers are working with trade unions, health organisations and other professional bodies to gain access to individuals willing to participate in research studies. It would be easier to work with employers. Some employers are beginning to participate actively in the research process (e.g. IRS, 1999). How to draw employers into the research should perhaps be a methodological consideration. Achieving participation may require a more subtle approach (such as not naming "bullying") and a flexible partnership attitude on the part of both the researcher and the employer.

The complexity of bullying at work as a field of study has emerged in recent years. Of credit to those who work in this area is their willingness to acknowledge, and their efforts to embrace, that complexity. We suggest that in the near future there may be a need for a review of our terminology so that definitions can be tightened for those who work in different contexts that are currently grouped under the heading "bullying at work". If this does happen, it is crucial that we do not lose sight of each other or the topic that we study.

Bullying at work appears to have evolved to the point that it has a name. Almost as that has occurred, it would appear that we might just have outgrown our name!Those whose primary interests lie in other fields such as conflict, attribution, management styles and organisational culture are beginning to apply their focus to bullying at work as an application area. It is therefore possible that the study of bullying at work may shatter into many pieces and disappear into a variety of other fields.

We would consider such a process of fragmentation regrettable. All the papers in this special issue evidence writers making valuable links between conceptual schools. We would suggest that it is through the use of academic theory (as a servant) in the study of bullying at work (rather than the other way round) that our real knowledge will be enhanced.

References
Archer, D. (1999), “Exploring ‘bullying’ culture in the para-military organisation”, International Journal of Manpower, Vol. 20 Nos 1/2.
Einarsen, S. (1999), “The nature and causes of bullying at work”, International Journal of Manpower, Vol. 20 Nos 1/2.
Fitzgerald, L.F. and Shullman, S. (1993), “Sexual harassment: a research analysis and agenda for the 1990’s”, Journal of Vocational Behaviour, Vol. 42, pp. 5-27.
Hoel, H., Rayner, C. and Cooper, C.L. (1999), “Workplace bullying”, International Review of Industrial and Organizational Psychology, Vol. 14, pp. 189-230.
Hofstede, G. (1980), Culture’s Consequences, Sage, Beverly Hills, CA .
IRS (1999), “Research on ‘destructive conflict’ at work”, Employee Health Bulletin, IRS Employment Review, No. 673, February, p. 5.
Keashly, L. (1998), “Emotional abuse in the workplace: conceptual and empirical issues”, Journal of Emotional Abuse, Vol. 1, pp. 85-117.
Keashly, L., Trott, V. and MacLean, L.M. (1994), “Abusive behaviour in the workplace: a preliminary investigation”, Violence and Victims, Vol. 9 No. 4, pp. 341-57.
Lazarus, R.S. (1976), Patterns of Adjustment, McGraw-Hill, New York, NY .
Leymann, H. (1990), “Mobbing and psychological terror at workplaces”, Violence and Victims, Vol. 5, pp. 119-25.
Leymann, H. (1996), “The content and development of mobbing at work”, European Journal of Work and Organizational Psychology, Vol. 5 No. 2, pp. 165-84.
Liefooghe, A.P.D. and Olafsson, R. (1999), “‘Scientists’ and ‘amateurs’: mapping the bullying domain”, International Journal of Manpower, Vol. 20 Nos 1/2.
Niedl, K. (1996), “Mobbing and wellbeing: economic and personnel development implications”, European Journal of Work and Organizational Psychology, Vol. 5 No. 2, pp. 239-49.
Rayner, C. (1997)), “The incidence of workplace bullying”, Journal of Community and Applied Social Psychology, Vol. 7 No. 3, pp. 199-208.
Rayner, C. (1999)), “From research to implementation: finding leverage for prevention”, International Journal of Manpower, Vol. 20 Nos 1/2.
Sheehan, M. (1999), “Workplace bullying: responding with some emotional intelligence”, International Journal of Manpower, Vol. 20 Nos 1/2.
Sheehan, M., Barker, M. and Rayner, C. (1999), “Applying strategies for dealing with workplace bullying”, International Journal of Manpower, Vol. 20 Nos 1/2.
Vartia, M. (1996), “The sources of bullying : psychological work environment and organizational climate”, European Journal of Work and Organizational Psychology, Vol. 5 No. 2, pp. 203-14.
Zapf, D., Knorz, C. and Kulla, M. (1996), “On the relationship between mobbing factors, job content, social work environment, and health outcomes”, European Journal of Work and Organizational Psychology, Vol. 5 No. 2, pp. 215-37.

Copyright © 1999 MCB. All rights reserved International Journal of Manpower, Vol 20 Issue 1/2 Date 1999 ISSN 0143-7720