SEARCH

SEARCH BY CITATION

Keywords:

  • occupational health;
  • personality;
  • violence

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

Aims and objectives.  To extend a model of the antecedents of workplace bullying to apply to a wider range of types of workplace aggression, including bullying and several types of violence, among nurses.

Background.  Research that has focused on workplace bullying has found that the Demand-Control-Support model, negative affectivity and certain demographic factors play important roles as antecedents of bullying.

Design.  A cross-sectional design.

Methods.  A validated questionnaire was sent to the work addresses of all nursing and midwifery staff in a medium-to-large hospital in Australia. A total of 273 nurses and midwives returned their completed questionnaires. Ordinal regressions were conducted to assess the antecedents of workplace aggression across bullying and violence.

Results.  Aspects of the Demand-Control-Support model and job tenure significantly predicted particular forms of violence, while negative affectivity and work schedule were significant for bullying.

Conclusions.  The patterns of the results suggest key mechanisms that characterise certain forms of violence and distinguish between bullying and types of violence across the range of workplace aggression. In particular, oppression and exposure appear to differentiate types of workplace violence.

Relevance to clinical practice.  The study suggests ways in which nursing and hospital managers may act to reduce the likelihood of certain forms of aggression, particularly violence, from occurring.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

Workplace aggression is a serious problem within the nursing occupation. High rates of workplace aggression, including bullying and violence, are frequently experienced by nursing staff (e.g. see Farrell et al. 2006, O’Connell et al. 2000) and linked to a range of negative consequences (e.g. Nabb 2000, Rowe & Sherlock 2005). Nurses working in hospital settings are argued to be an oppressed group, whereby these settings are influenced by medical hierarchies through which nurses lack power and control (Roberts 1983, Roberts et al. 2009), thus highlighting the vulnerability of nurses to workplace aggression (Freire 1971).

Workplace aggression research specifically examining nursing staff has tended to focus on the antecedents of workplace bullying (e.g. Hutchinson et al. 2008, 2010). Other forms of workplace aggression, such as threat of violence or even violence itself, can also occur in the workplace, yet the antecedents of these forms are rarely investigated within nursing. Previous research that has examined the antecedents of workplace bullying for nurses and other occupational groups has found that aspects such as job demands, job control and social support can play important roles (Baillien et al. 2011). Research has also found that individual factors, such as negative affectivity (NA; i.e. an individual’s level of pervasive negative emotionality and self-concept; Watson & Clark 1984) and certain demographic factors, can have an important impact (e.g. Hansen et al. 2006). Thus, it might be beneficial to extend the approach used within workplace bullying research by examining the utility of applying work stress/bullying models to a range of workplace aggression that includes bullying and types of violence. Subsequently, the overall aim of this study was to test this multidimensional model of the antecedents of workplace aggression across several forms of bullying and violent behaviours.

Background

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

Oppressed group behaviours within the nursing profession are often referred to as ‘horizontal violence’ and can take the form of a type of workplace aggression. Freire (1971) theorised that oppressed group behaviours can occur when the powerless are submissive and silent in confrontation with authority, and consequently, fear and low self-esteem result that lead to anger and aggressive behaviours internally towards one’s own group members. Such workplace aggression behaviours have been documented among nurses and have been linked to negative aspects of the nurse workplace (Roberts et al. 2009).

Workplace aggression, as perceived by the victim, can manifest in a variety of forms, including those that are bullying and violence related. Workplace aggression is therefore the overarching concept that encompasses acts of bullying and violence. Although researchers are yet to agree on uniform definitions of these types of aggression, there are consistent features across definitions for bullying and violence, key features of which are reviewed here.

The nature of workplace bullying can be both psychological and/or physical; however, many researchers tend to agree that workplace bullying is primarily psychological in nature, including both perceived and/or actual psychological harm (Einarsen et al. 2011). Sources of bullying are typically internal from within the organisation, such as co-workers and/or supervisors. Similarly, the frequency of bullying does not include single and isolated events, but rather repeated and persistent negative behaviours towards the employee being bullied (Einarsen et al. 2011). Given these features of bullying, the following definition has been put forth: bullying as defined as an individual’s perception of receiving and having difficulty defending, the negative actions from one or several others that is persistent over a period of time (Hoel & Cooper 2000).

In contrast to workplace bullying, the nature of workplace violence is primarily focused on perceived threat or actual physical harm; however, violence can also include acts that are psychological. For hospital nurses, multiple sources of violence exist, including violent acts from co-workers and supervisors, as well as violent acts from patients, family and visitors of patients (Hesketh et al. 2003). Thus, violence sources may be internal or external to the organisation, respectively. Examples of types of violence include physical assault, threat of assault, emotional abuse and verbal sexual harassment (Hesketh et al. 2003). A definition of violence includes examples of the types of violence, whereby physical assault might include being spit on, bitten, hit and/or pushed, threat of assault can include verbal and/or written threats intending to harm, emotional abuse involves insults, gestures, humiliation before the work team and/or coercion, and finally, verbal sexual harassment comprises unwanted intimate questions and/or remarks of a sexual nature (Hesketh et al. 2003). Lastly, both single and repeated acts of violence are encompassed in the definition of violence.

The main areas of overlap between these forms of workplace aggression include the nature of bullying and violence comprising both psychological and physical acts. Although, the primary nature of these acts as either psychological for bullying or physical for violence is also a key distinction between the two. Further, a distinguishing factor is the frequency of negative acts as repeated for bullying and single or repeated for violence. The sources of bullying are also distinct from violence, with bullying more commonly from internal sources (i.e. supervisors and co-workers) and violence potentially sourced from internal or external sources (i.e. the patients or the patient’s family and friends).

Given the distinctions of what constitutes bullying and violence behaviours at work, it might be worthwhile to examine both forms of workplace aggression together to provide greater insight into the similarity or dissimilarity of the underlying factors leading to each. Previous research that has examined nurses has tended to focus on the antecedents of one type of workplace aggression, most commonly workplace bullying (e.g. Hutchinson et al. 2008, 2010), with studies rarely examining both types of aggression simultaneously. Therefore, the analytic approach used in previous research with workplace bullying in nurses and other occupational groups might provide a valuable source of information in extending our understanding of the antecedents of workplace aggression as a whole to include acts of bullying and violence.

Research that has examined the antecedents of workplace bullying has found that the Demand-Control-Support (DCS) model is a useful framework (e.g. Baillien et al. 2011, Tuckey et al. 2009). The DCS model is one of the most widely used frameworks within occupational stress research (De Lange et al. 2003). Originally, Karasek (1979) developed the model with two dimensions consisting of only job demands and control variables. In this two-dimensional model, Karasek postulated that an interaction of job demands with control can predict psychological and physical strain. Accordingly, high-strain jobs are said to consist of high job demands and low levels of job control. In later research, the model was expanded to include social support, forming the DCS model (Johnson & Hall 1988). Johnson and Hall (1988) found a buffering effect of social support, whereby high-strain jobs were those characterised by high demands, low control and/or social support.

The DCS model has been tested and validated for various occupational groups and outcomes, such as job satisfaction, organisational commitment, psychological distress and well-being (Taris et al. 2001, Mikkelsen et al. 2005, Noblet et al. 2006), including workplace bullying as a stressor (e.g. Baillien et al. 2011, Tuckey et al. 2009). More specifically, Baillien et al. (2011) tested the demand and control model in a sample of blue-collar workers (i.e. employees who perform manual labour, such as manufacturing) and found significant interaction effects between demand and bullying in conditions of low control, as well as strong main effects of demand itself. Further, Tuckey et al. (2009) examined the moderating effects of job control and social support resources against job demands for workplace bullying in a sample police officers. Increased levels of bullying were found to be associated with increased job demands and decreases in the control and support resources. Thus, these studies highlight the ability of the DCS model in being applicable to a form of workplace aggression. However, there is limited research that explores these relationships in nurses across other forms of workplace aggression, including both bullying and violence.

Further, there is some evidence to suggest that particular personality traits may also be important to consider when examining the antecedents to workplace aggression. More specifically, the research on workplace bullying has revealed that NA can play a significant role (Mikkelsen & Einarsen 2002, Hansen et al. 2006). One explanation of the relationship of NA as a potential antecedent of bullying at work is the view that NA may act as a potential perceptual bias, whereby individuals with high levels of NA perceive behaviours as more personal than they actually are, and consequently report greater instances of workplace bullying (Mikkelsen & Einarsen 2002). Mikkelsen and Einarsen (2002) discuss this process as a cycle through which an individual is exposed to interpersonal conflict that increases their distress levels, and while distressed, the individual interprets the conflict as negative and personal, leading to increased experiences of negative emotions.

Another possible explanation of NA as a potential antecedent of workplace bullying is derived from the closely related field of work stress, known as the stressor creator mechanism (Spector et al. 2000). According to this mechanism, individuals with high NA may create or enact negative circumstances that lead to job stress. For example, these individuals may be more likely to get into interpersonal conflicts at work and/or poorly perform on their job (Spector et al. 2000). In line with this explanation, NA may create stressors, such as workplace bullying.

Given the previous findings of an association between NA and workplace bullying, as well as these proposed underpinnings, it might be plausible to suggest that NA has the same or similar underlying mechanisms for other forms of aggression experienced by nurses, such as workplace violence. Limited research has tested NA this way, where NA may be a possible antecedent to workplace aggression in many of its forms for nursing staff.

Finally, another aspect that may be worthy of consideration as a potential antecedent to workplace aggression are certain demographical factors whereby these factors impact the occurrence or perception of workplace aggression acts. Some research that explores bullying separately in nurses has revealed that tenure level and work schedule may alter reports of workplace bullying (e.g. Duncan et al. 2001). Thus, it might also be valuable to expand tests of the antecedents to workplace aggression in nursing to include potential demographical effects as well.

The overall aim of the present study is to investigate the antecedents of workplace aggression in nursing staff, by applying what we already know about workplace bullying to other forms of aggression, including workplace violence. Based on the findings from workplace bullying research, several hypotheses are put forth for all workplace aggression types. The first hypothesis relates to the DCS model, whereby high job demands, low job control and low social support are hypothesised to be associated with reports of workplace aggression. It is also hypothesised that higher levels of NA will be associated with reports of workplace aggression. Finally, it is hypothesised that the demographical factors of tenure and work schedule will significantly impact reports of workplace aggression.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

Sample and Procedure

The sample comprised nurses and midwives working together across all wards at a medium-to-large hospital within Australia. A cross-sectional design was used. Questionnaire packs were sent to the work addresses of all nursing and midwifery staff across the hospital. Participation was both voluntary and anonymous. Ethics approval was sought and obtained from the university and healthcare organisation prior to the questionnaire mail out.

A total of 273 nurses and midwives returned their completed questionnaires. The overall response rate for the sample was 37·1%. To summarise the demographic characteristics of the sample, all respondents were women (= 267, 100%), with the majority aged between 50 and 54 years old (= 61, 22·5%), followed by those over 50 years old (= 48, 17·7%). Most respondents had worked in their job for nine years or less (= 187, 69·0%), followed by those who had worked for 20 years or more (= 63, 23·7%). A large portion of respondents were employed on a part-time basis (= 164, 60·1%), with full time (= 75, 27·5%) and casual staff (= 33, 12·1%) comprising smaller proportions. Exactly half of the employees indicating their most recent work shifts had worked morning shifts (= 94, 50·0%), followed by night shifts (= 36, 19·1%). The majority of respondents reported that they had completed some form of tertiary education, including tertiary degrees/diplomas (= 96, 36·9%) and postgraduate studies (= 129, 49·6%).

Instruments

The frequency of bullying was measured using a single item taken from Hoel and Cooper (2000). Respondents were asked to indicate whether or not they had experienced bullying at work within the last six months. Respondents were asked to use the following definition of workplace bullying when responding, ‘Bullying is defined as a situation where individuals persistently, over a period of time, perceive themselves to be on the receiving end of negative actions from one or several persons, in a situation where the target of bullying has difficulty in defending him or herself against these actions. For the purposes of this survey, one-off incidents are not referred to as bullying’. The response options on the item were ‘no’, ‘yes, very rarely’, ‘yes, now and then’, ‘yes, several times a month’, ‘yes, several times a week’ and ‘yes, almost daily’. This scale by Hoel and Cooper is a well-validated measure that has been supported by comparisons with a standardised 22-item validated scale of bullying: the Negative Acts Questionnaire-Revised (Einarsen et al. 2009).

The frequency of violence was measured using an adapted version of the scale developed by Hesketh et al. (2003). Respondents were ask to indicate whether in the last five work shifts, they had experienced any of the following types of violence: physical assault (e.g. being spit on, bitten, hit and/or pushed), threat of assault (e.g. verbal and/or written threats intending to harm), emotional abuse (i.e. insults, gestures, humiliation before the work team and/or coercion) and verbal sexual harassment (i.e. repeated, unwanted intimate questions and/or remarks of a sexual nature). The response options for each of the items included ‘never’, ‘1 time’, ‘2 times’ and ‘3 or more times’. Each item also asked for the source of violence: patient, visitor/family member, co-worker and supervisor. These sources of violence were collapsed to form subscales of internal and external violence across types of violence. Internal violence included violent acts experienced from a supervisor or co-worker, while external violence was from a patient or visitor/family member.

Job demands were measured using a scale developed by Caplan et al. (1980). This scale asked respondents about their perceptions of the levels of physical and psychological demands their job places on them. There were 11 items on the scale, rated on a five-point Likert scale, ranging from ‘rarely’ to ‘very often’. The Cronbach’s alpha of job demands in this study was .89.

Job control was measured using a scale developed by Karasek (1985). This scale assesses perceptions of control one perceives they have over the demands of their job. The scale has nine items, rated on a five-point Likert scale, ranging from ‘strongly disagree’ to ‘strongly agree’. The Cronbach’s alpha of job control in this study was .73.

Social support was measured using a scale developed by Caplan et al. (1980). There were four items on the scale requesting that respondents indicate the level of support they believe to receive from their immediate supervisor, work colleagues and outside work. These three responses for each item were used to form three subscales of social support: supervisor support, co-worker support and outside work support. Each item was rated on the scale using a five-point Likert scale ranging from ‘do not have any such person’ to ‘very much’. The Cronbach’s alphas for supervisor support, co-worker support and outside work support in the present study were .88, .80 and .81, respectively.

NA was measured using the NA scale from the Positive and Negative Affect Schedule (PANAS) by Watson et al. (1988). Respondents were asked the extent to which they experienced particular negative emotions within the past week. For example, negative emotions of distress and hostility were queried. The scale has 10 items with response options rated on a five-point Likert scale from ‘very slightly or not at all’ to ‘very much’. The Cronbach’s alpha for negative affectivity in this study was .77.

Demographical variables were included in the study’s analyses, namely tenure and work schedule. Tenure categories included ‘9 years or less’, ‘10–14 years’, ‘15–19 years’, ‘20–24 years’ and ’25 years or more’. Work schedule categories included ‘morning’, ‘afternoon’, ‘night’ and ‘other’ shifts.

Data Analyses

Statistical analyses included conducting descriptive statistics and ordinal regressions using Predictive Analytics SoftWare (pasw) 17.0 (SPSS Inc 2010). Ordinal regressions assessed the antecedents of bullying and all forms of violence and were conducted following Hair et al. (1995). The antecedents that were included in these regressions were the DCS variables (i.e. job demands, job control, supervisor support, co-worker support and outside work support), NA and the demographic variables of tenure (i.e. 9 years or less, 10 to 14 years, 15 to 19 years, 20 to 24 years and 25 or more years) and work schedule (i.e. morning shifts, afternoon shifts, night shifts and other shifts). Prior to conducting these analyses, preliminary checks were conducted to ensure that there were no violations of the assumptions.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

Frequency of Bullying and Violence

To ensure adequate sample sizes across groups for statistical purposes, the bullying responses were collapsed into those who responded ‘no’, ‘yes, rarely’ and ‘yes, frequently’. Similarly, the violence response options were collapsed into ‘no’, ‘yes, rarely’ and ‘yes, frequently’. With these groupings, the frequency of bullying and violence, including all violence types, is presented in Table 1.

Table 1.   Frequencies and percentages of bullying and violence responses
Workplace aggressionResponse options
NoYes, rarelyYes, frequently
n (%)n (%)n (%)
Bullying167 (61·9)53 (19·6)50 (18·5)
Violence
 External physical assault263 (97·0)6 (2·2)2 (0·7)
 Internal physical assault266 (97·8)3 (1·1)3 (1·1)
 External threat of assault243 (90·0)20 (7·4)7 (2·6)
 Internal threat of assault265 (98·9)1 (0·4)2 (0·7)
 External emotional abuse216 (80·9)35 (13·1)16 (6·0)
 Internal emotional abuse199 (74·3)47 (17·5)22 (8·2)
 External verbal sexual harassment 264 (97·8)5 (1·9)1 (0·4)
 Internal verbal sexual harassment267 (98·9)0 (0·0)3 (1·1)

Table 1 highlights the most frequent report of workplace aggression within the ‘yes, frequently’ responses of bullying (= 50, 18·5%), followed by violence that was emotional abuse across internal (= 22, 8·2%) and external (= 16, 6%) sources. Within the ‘yes, rarely’ responses, once again bullying (= 53, 19·6%), followed by emotional abuse across internal (= 47, 17·5%) and external (= 35, 13·1%) sources, was the most frequent report of aggression.

Examining the Antecedents of Bullying and Violence

The results of the ordinal regressions are detailed in Table 2. For bullying, two significant antecedents were identified, namely there was a positive relationship between NA and bullying (β = 0·08, < 0·001), whereby high levels of NA were associated with reports of bullying acts. Also, there was a positive relationship between the work schedule of morning shift and bullying (β = 0·66, < 0·05), with morning shift workers more likely to experience bullying than other shift workers.

Table 2.   Results of the ordinal regressions for the antecedents of workplace aggression
Independent variablesBullyingViolence
External physical assaultInternal physical assaultExternal threat of assaultInternal threat of assaultExternal emotional abuseInternal emotional abuseExternal verbal sexual harassmentInternal verbal sexual harassment
  1. *< 0·05; **< 0·01; ***< 0·001.

Job control−0·030·01−0·06−0·02−0·61−0·02−0·07*−0·25*−0·24*
Demand0·030·020·070·07*0·280·08**0·01−0·130·10
Supervisor support−0·05−0·19−0·55−0·06−0·550·08−0·15**−0·07−0·17
Co-worker support−0·070·270·25−0·040·290·01−0·15**−0·040·03
Outside work support−0·030·23−0·060·031·72−0·000·050·160·19
Negative affect0·08***0·030·22−0·01−0·000·030·010·200·03
Job tenure
 9 years or less0·4615·3216·18***19·03***2·620·12−0·6215·1715·48
 10–14 years−0·290·1918·97***18·87***1·46−0·63−0·4915·590·12
 15–19 years0·9416·3319·6719·98***−8·670·320·510·820·84
 20–24 years−0·03−0·06−1·9419·044·641·030·1717·2617·41
 25+ years
Work schedule
 Morning shift0·66*0·2722·15−0·207·83−0·28−0·230·090·45
 Afternoon shift−0·3315·852·810·035·16−0·09−0·3516·5915·79
 Night shift−0·29−0·4819·40−0·881·63−0·46−0·66−0·2315·44

The results for the violence types revealed that internal physical assault was significantly associated with several ranges of job tenure. More specifically, internal physical assault was more likely for nurses and midwives employed for 9 years or less (β = 16·18, < 0·001) or between 10–14 years (β = 18·97, < 0·001), when compared with nurses and midwives with a job tenure of 25 years or more.

Significant relationships were observed for demand and several types of job tenure predicting external threat of assault. Reports of high demand were associated with reports of external threat of assault. Further, reports of external threat of assault were more common among nurses and midwives employed for 9 years or less (β = 19·03, < 0·001), between 10–14 years (β = 18·87, < 0·001) and 15–19 years (β = 19·98, < 0·001), in comparison with a job tenure of at least 25 years.

Only one significant relationship was identified for external emotional abuse, with high levels of demand related to high levels of external emotional abuse (β = 0·08, < 0·01). In contrast, the results for internal emotional abuse revealed significant relationships with job control, supervisor support and co-worker support. The direction of these relationships indicated that low job control (β = −0·07, < 0·05), low supervisor support (β = −0·15, < 0·01) and low co-worker support (β = −0·15, < 0·01) were associated with reports of internal emotional abuse. There were no significant predictors for external physical assault or internal threat of assault. In terms of the external and internal verbal sexual assault types, job control was significantly associated with negative relationships observed for both types of violence through which low job control was associated with greater reports of external (β = −0·25, < 0·05) and internal (β = −0·24, < 0·05) verbal sexual assault.

To examine the extent to which the model variation was minimised by the inclusion of the independent variables, the likelihood ratio ( inline image ), analogous to the R2 statistic employed in linear regression, was derived through a two-step process outlined in Menard (2002). Applying this process determines that 12·8% of the proportion of variance in bullying was accounted for by the significant independent variables. The regression models also explained 62% of the variance of internal physical assault, 38% of the variance of internal verbal sexual harassment and 15·4% of the variance of internal emotional abuse. For external verbal sexual harassment, logistic regression confirmed the aforementioned results, with yes and rarely collapsed together, and had 36% of the variance predicted. In contrast, only 7% of the variance in external threat of assault and 6·9% of the variance in external emotional abuse were explained by the ordinal regression models.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

This study investigated the antecedents of workplace aggression in nursing staff, by applying a model from the workplace bullying and stress literature to other forms of aggression, particularly workplace violence. The results found high levels of bullying (18·5% reported occasional or frequent bullying) and concerning levels of emotional abuse – both internal and external, as well as external threat of assault (10% reported rare or higher). Together, these prevalence levels show worryingly high rates of workplace aggression, confirming prior research on aggression in nursing (O’Connell et al. 2000, Farrell et al. 2006), particularly given the likely associated negative consequences (e.g. Nabb 2000, Rowe & Sherlock 2005). That is, this study’s results confirm that workplace aggression is a serious problem within nursing.

Contrary to previous research (Tuckey et al. 2009, Baillien et al. 2011), the DCS variables did not predict bullying. However, demand did predict external threat of assault and external emotional abuse, both of which are non-physical forms of aggression coming from non-employees.

Low levels of job control were significantly associated with internal emotional abuse and both internal and external verbal sexual harassment. These results for job control appear to highlight that the argument that nurses are an oppressed group with reduced levels of power (Roberts 1983, Roberts et al. 2009) is particularly predictive of verbal sexual harassment (internal and external) and internal emotional abuse for those nurses with low job control. These results suggest that, in terms of workplace characteristics, job control is an indicator of the more vulnerable nursing staff. Similarly, while the stress literature emphasises the potential buffering affects of social support (Johnson & Hall 1988), in the context of predicting forms of aggression, lower levels of social support acted to exacerbate or facilitate internal emotional abuse. That is, the potential for internal emotional abuse appears to be enabled under conditions of social isolation, possibly because of the mechanisms of oppressed group behaviours (Freire 1971) or social out-group processes.

A quite surprising finding was the relative lack of significant results for NA. The results reported here support prior research on the importance of NA’s role in workplace bullying (e.g. per Hansen et al. 2006, Mikkelsen & Einarsen 2002). NA significantly predicting bullying suggests that NA may be acting through the stressor-creation mechanism (Spector et al. 2000) where individuals with high NA may be more likely to get into bullying-prone situations at work. However, the finding that NA did not predict any of the forms of violence not only distinguishes bullying from violence but also weakens the likelihood that the individual can ‘attract’ violence. That is, our findings appear to highlight the extra-individual nature of violence with violence being inflicted upon the victim. This result also contrasts the role of NA in bullying where the individual has been proposed to potentially be an active contributor to the bullying ‘cycle’ (Mikkelsen & Einarsen 2002) where high-NA individuals are more likely to be distressed and the distressed high-NA individuals are more likely to interpret any workplace conflict as negative and personal, leading to increased perceptions of negative outcomes and so on.

The results for the demographic variables provide some interesting caveats to the results. When the results of the aforementioned analyses were discussed via qualitative feedback with the project’s representatives from the studied hospital, the discussants noted (in a similar manner to Rowe & Sherlock 2005) that the less qualified staff are more likely to be given the least desirable work on the morning shifts in this hospital, which may have been interpreted as bullying by those staff and that this informal practice may be the cause of the significant morning shift results. For internal physical assault and external threat of assault, the other significant demographic variable was tenure. The pattern of loadings appears to indicate a coalface or ‘exposure’ effect where the nurses most likely to suffer these two forms of violence are those that have the most exposure to them and are working on the hospital equivalent to a coalface. That is, the main indication from the tenure results is that more senior staff are less likely to be the victims of those forms of violence, where nurses with 15 or more years of experience are less likely to be in roles that could expose them to internal physical assault, and similarly, those nurses with more than 20 years experience are less likely to have contact with external people.

Finally, there were no significant predictors for external physical assault and internal threat of assault. These non-results show that if these types of violence are systematically determined, then they may not be determined by variables measured in this study. For example, external physical assault may be more a product of the patient’s degree of drunkenness and by the time of night. Alternatively, the occurrences may have a larger degree of randomness or be so uncommon (e.g. because of social norms) that they rarely occur. For example, the internal threat of assault had the lowest rate of occurrence of all the types of violence. However, it is notable that internal physical assault (i.e. actual physical assault) was more prevalent than threatened internal physical assault.

Workplace aggression can manifest in a variety of forms, whether bullying or violence related. This study’s results give some indication of key characterisations of the various types of workplace aggression that may be worth exploring further in future research. An initial delineation between bullying and the other types of violence was that NA was only significant for bullying and was not significant for any of the other types of aggression. This result highlights the individual-involved nature of bullying where the individual’s perceptions play a role in the perception of bullying having occurred. The role of NA is potentially further enhanced because of the definition of bullying used in this study applying the requirement that the bullying was repeated and not one-off, thereby allowing multiple opportunities for high-NA individuals to begin a negative cycle (e.g. Mikkelsen & Einarsen 2002).

Perhaps the most useful grouping of the types of violence covered in this study was the classification as either internally sourced (i.e. from co-workers and supervisors) or externally sourced (e.g. from patients and/or the patients’ families). Demand being a significant predictor for external threat of assault and external emotional abuse may indicate a mechanism where the sheer busyness of the nurses under high demands may have been perceived by the external party as a lack of attention (whether real or not) from the nurse, resulting in the external party’s violent behaviour. The implicit opportunity in this finding is that management can act to address this cause and pay particular attention to managing the work demands of nurses, especially for nurses in areas with substantial external contact and ‘exposure’.

The finding that low job control was associated with verbal sexual harassment, whether internal or external, suggests that control over work may be a determinant of workplace situations where violence-as-oppression may occur. These results apply the oppression theory of aggression (Roberts 1983, Roberts et al. 2009) to the modern nursing workplace by proposing that sexual harassment may be the form of violence that is most likely to occur in a situation of oppression, particularly in the context of a hospital dominated by medical hierarchies. That is, oppressed group behaviours are theorised (Freire 1971) as being facilitated when the powerless are submissive and silent in confrontation with authority, indicated by low levels of job control in this study, and consequently, fear and low self-esteem result. This study’s results show that the oppressed group, in the case of hospital nurses, may then be the victims of sexual harassment. Similarly, within the internal forms of violence, this study’s results suggest that internal emotional abuse may be enacted through oppressed group behaviour (Freire 1971) or social out-group processes, with a further slight oppression effect for nurses with less job control being more at risk.

Limitations

The results of the study are interpreted in the light of the limitations of cross-sectional study designs and self-report measures. The cross-sectional nature of the study impacts the ability of the authors to draw conclusions about the causality of the antecedents of workplace aggression. Future research that adopts longitudinal study designs to test these types of effects is recommended. Further, the self-reported data present issues associated with common method variance and common source bias. However, it is important to note that to study bullying and violence from the victim’s perspective, self-reported measures are most appropriate (e.g. see Goffin & Gellatly 2001). Further, the prospective time scales of the bullying and violence items differ, with bullying measured over six months and violence in the previous five working shifts. Previously validated scales were chosen for this study, namely those by Hesketh et al. (2003) for the violence measure and Hoel and Cooper (2000) for the bullying measure. Future research that devises such scales could benefit from attempting to have a consistent time focus to improve comparability. Finally, the response rate of the study was relatively low, with 37·1% of nurses and midwives participating in the study. Although significant findings existed, the conclusions drawn should be considered in the context of this response rate.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

This study extends a model used in workplace bullying research to a broader range of aggression, across several types of violence for nurses. The patterns of the results suggest some very interesting mechanisms that help to delineate the various forms of violence, as well as highlighting some mechanisms that may be common to various violence types. This study is one of very few studies that have investigated workplace characteristics that impact on the occurrence of a range of types of aggression, including bullying and various types of violence, by whether internally or externally sourced. The predictors also helped to differentiate the types of workplace violence by highlighting the likely mechanisms of specific types of violence. Although the individual’s perceptions are integrally involved with the bullying process, many of the violence types had an extra-individual quality that means that characteristics of the workplace, rather than individual perceptions, may play a role. In particular, the apparent connection between oppression indicators and verbal sexual harassment is very concerning and, as often implicit in occupational health and safety policies, the nature of other violence types appeared to be more a matter of exposure.

Relevance to clinical practice

To support the recruitment and retention of nurses in the workforce, it is important to determine what factors might predict the occurrence of workplace aggression, given that high rates of aggression within this professional field can potentially deter prospective nurses from entering, or drive current nurses out of, the workforce. Investigating a workplace model across a variety of types of violence increases our understanding of workplace-related areas to target and intervene, with the aim of effectively reducing the occurrence, as well as associated negative consequences, of workplace aggression. Using an empirically supported workplace model, the results of this study suggest ways in which nursing and hospital managers may act to reduce the likelihood of certain forms of aggression, particularly violence, from taking place, especially in terms of nursing staff with low levels of job control and managing interactions on morning shifts.

Contributions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

Study design: JR; data collection and analysis: JR, DD and manuscript preparation: JR, DD.

Conflict of interest

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

All authors confirm that there are no actual or potential conflicts of interest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References
  • Baillien E, Rodríguez-Muñoz A, de Witte H, Notelaers G & Moreno-Jiménez B (2011) The Demand–Control model and target’s reports of bullying at work: a test within Spanish and Belgian blue-collar workers. European Journal of Work and Organizational Psychology 20, 157177.
  • Caplan R, Cobb S, French J Jr, Harrison R & Pinneau S (1980) Job Demands and Worker Health: Main Effects and Occupational Differences. The Institute for Social Research, Ann Arbour, MI.
  • De Lange AH, Taris TW, Kompier MAJ, Houtman ILD & Bongers PM (2003) “The very best of the millennium”: longitudinal research and the demand-control-(support) model. Journal of Occupational Health Psychology 8, 282305.
  • Duncan S, Hyndman K, Estabrooks C, Hesketh K, Humphrey C, Wong J, Acorn S & Giovanetti P (2001) Nurses’ experience of violence in Alberta and British Columbia hospitals. The Canadian Journal of Nursing Research 32, 5778.
  • Einarsen S, Hoel H & Notelaers G (2009) Measuring exposure to bullying and harassment at work: validity, factor structure and psychometric properties of the negative acts questionnaire-revised. Work & Stress 23, 2444.
  • Einarsen S, Hoel H, Zapf D & Cooper CL (2011) The concept of bullying and harrassment at work: the European tradition. In Bullying and Harrassment in the Workplace: Developments in Theory, Research and Practice (Einarsen S, Hoel H, Zapf D & Cooper CL eds). CRC Press, Boca Raton, FL, pp. 339.
  • Farrell G, Bobrowski C & Bobrowski P (2006) Scoping workplace aggression in nursing: findings from an Australian study. Journal of Advanced Nursing 55, 778787.
  • Freire P (1971) Pedagogy of the Oppressed. Herder & Herder, New York, NY.
  • Goffin RD & Gellatly IR (2001) A multi-rater assessment of organizational commitment: are self-report measures biased? Journal of Organizational Behavior 22, 437451.
  • Hair JF Jr, Anderson RE, Tatham RL & Black WC (1995) Multivariate Data Analysis with Readings. Prentice Hall, Englewood Cliffs, NJ.
  • Hansen ÅM, Hogh A, Persson R, Karlson B, Garde AH & Ørbæk P (2006) Bullying at work, health outcomes, and physiological stress response. Journal of Psychosomatic Research 60, 6372.
  • Hesketh KL, Duncan SM, Estabrooks CA, Reimer MA, Giovannetti P, Hyndman K & Acorn S (2003) Workplace violence in Alberta and British Columbia hospitals. Health Policy 63, 311321.
  • Hoel H & Cooper CL. (2000). Destructive conflict and bullying at work. Unpublished report, University of Manchester Institute of Science and Technology (UMIST).
  • Hutchinson M, Jackson D, Wilkes L & Vickers M (2008) A new model of bullying in the nursing workplace: organizational characteristics as critical antecedents. Advances in Nursing Science 31, 6071.
  • Hutchinson M, Wilkes L, Jackson D & Vickers M (2010) Integrating individual, work group and organizational factors: testing a multidimensional model of bullying in the nursing workplace. Journal of Nursing Management 18, 173181.
  • Johnson JV & Hall EM (1988) Job strain, work place social support, and cardiovascular disease: a cross-sectional study of a random sample of the Swedish working population. American Journal of Public Health 78, 1336.
  • Karasek R. (1979). Job demands, job decision latitude, and mental strain: Implications for job redesign. Administrative Science Quarterly 24, 285308.
  • Karasek R (1985) Job content questionnaire and user’s guide. Department of Industrial and Systems Engineering, LA.
  • Menard S (2002) Applied Logistic Regression analysis. Sage Publications, Thousand Oaks, CA.
  • Mikkelsen EG & Einarsen S (2002) Relationships between exposure to bullying at work and psychological and psychosomatic health complaints: the role of state negative affectivity and generalized self–efficacy. Scandinavian Journal of Psychology 43, 397405.
  • Mikkelsen A, Øgaard T & Landsbergis P (2005) The effects of new dimensions of psychological job demands and job control on active learning and occupational health. Work & Stress 19, 153175.
  • Nabb D (2000) Visitors’ violence: the serious effects of aggression on nurses and others. Nursing Standard 14, 3638.
  • Noblet AJ, McWilliams J, Teo STT & Rodwell JJ (2006) Work characteristics and employee outcomes in local government. The International Journal of Human Resource Management 17, 18041818.
  • O’Connell B, Young J, Brooks J, Hutchings J & Lofthouse J (2000) Nurses’ perceptions of the nature and frequency of aggression in general ward settings and high dependency areas. Journal of Clinical Nursing 9, 602610.
  • Roberts S (1983) Oppressed group behavior: implications for nursing. Advances in Nursing Science 5, 2130.
  • Roberts S, Demarco R & Griffin M (2009) The effect of oppressed group behaviours on the culture of the nursing workplace: a review of the evidence and interventions for change. Journal of Nursing Management 17, 288293.
  • Rowe MM & Sherlock H (2005) Stress and verbal abuse in nursing: Do burned out nurses eat their young? Journal of Nursing Management 13, 242248.
  • Spector PE, Zapf D, Chen PY & Frese M (2000) Why negative affectivity should not be controlled in job stress research: Don’t throw out the baby with the bath water. Journal of Organizational Behavior 21, 7995.
  • SPSS Inc. (2010) Predictive Analytics SoftWare (PASW) 17.0. IBM Corporation, Somers, NY.
  • Taris TW, Schreurs PJG & Van Iersel-Van Silfhout IJ (2001) Job stress, job strain, and psychological withdrawal among Dutch university staff: towards a dualprocess model for the effects of occupational stress. Work & Stress 15, 283296.
  • Tuckey MR, Dollard MF, Hosking PJ & Winefield AH (2009) Workplace bullying: the role of psychosocial work environment factors. International Journal of Stress Management 16, 215232.
  • Watson D & Clark LA (1984) Negative affectivity: the disposition to experience aversive emotional states. Psychological Bulletin 96, 465490.
  • Watson D, Clark LA & Tellegen A (1988) Development and validation of brief measures of positive and negative affect: the PANAS scales. Journal of Personality and Social Psychology 54, 10631070.

The Journal of Clinical Nursing (JCN) is an international, peer reviewed journal that aims to promote a high standard of clinically related scholarship which supports the practice and discipline of nursing.

For further information and full author guidelines, please visit JCN on the Wiley Online Library website: http://wileyonlinelibrary.com/journal/jocn

Reasons to submit your paper to JCN:

High-impact forum: one of the world’s most cited nursing journals and with an impact factor of 1·228 – ranked 23 of 85 within Thomson Reuters Journal Citation Report (Social Science – Nursing) in 2009.

One of the most read nursing journals in the world: over 1 million articles downloaded online per year and accessible in over 7000 libraries worldwide (including over 4000 in developing countries with free or low cost access).

Fast and easy online submission: online submission at http://mc.manuscriptcentral.com/jcnur.

Early View: rapid online publication (with doi for referencing) for accepted articles in final form, and fully citable.

Positive publishing experience: rapid double-blind peer review with constructive feedback.

Online Open: the option to make your article freely and openly accessible to non-subscribers upon publication in Wiley Online Library, as well as the option to deposit the article in your preferred archive.