Department of psychosocial science

Pesten op het werk wordt gedefinieerd als een gradueel escalerend proces. Via de gangbare metingen wordt dit proces echter onvoldoende gevat. Deze metingen vertonen tevens diverse inhoudelijke en methodologische tekortkomingen. In deze bijdrage wordt daarom een latente klassenanalyse uitgevoerd op data die met de Negative Acts Questionnaire werden verkregen ('objectieve' meting van pesten). De geanalyseerde data (N = 6175) laten toe zes clusters te onderscheiden. Deze zijn: de 'niet-gepesten' (35,3%), het 'beperkt werkgerelateerde kritiek'-cluster (27,7%), het 'beperkt negatief bejegenden'-cluster (16,5%), de 'soms gepesten' (9%), de 'werkgerelateerd gepesten' (8,3%) en de 'slachtoffers' (3,2%). Deze resultaten suggereren een cumulatief meetmodel voor pesten op het werk, waarbij het type pestgedragingen en hun intensiteit systematisch toenemen. Dit levert resultaten op die beter aansluiten bij de definitie van pesten op het werk. Ook de omvang van de slachtoffergroep sluit beter aan bij de percentages die internationaal gerapporteerd worden wanneer een subjectieve meetmethode wordt gebruikt.

The aim of this research is to help organisations and companies to develop a policy of counteracting stress at work.  The analysis is kept very simple.  We use analysis of variance to find out if belonging to a certain group makes a difference.  The result is an overview of the organisation’s position as opposed to a reference population on all the questioned angles (scales) which have quite good measurement properties.

Taking into account that the items in question are Mokken items, we only have to add up the items to get scalescores.  The scores of the scales are standardised from 0 to 100 where 0 means having nothing of a particular latent trait and 100 meaning completely applicable or fully experiencing the latent treat. 


An interesting point is that the author of the questionnaire proposes a second-order factoranalysis containing six factors : Psychological Demands, Variety, Possibilities to Control, Social Organisational, Satisfaction (well-being) and Tension.

This structure provides the starting point for this paper on multidimensional scaling.  First we want to see if the structure van Veldhoven proposes holds for the Belgian benchmark.  Secondly we are interested in the fact if being a blue collar worker, an employee, a nurse or a manager makes a difference.  For this reason we like to fit an individual differences scaling model for the data from the Belgian benchmark.  Thirdly we like to investigate if the data gathered in the Netherlands and Belgium by matching on key variables as age group, educational level, industry, being a manager or not and gender, behave differently.  This is also an invitation to use a individual differences scaling model.


Knowing that ALSCAL is supported by Forrest Young in SPSS minimises S-STRESS and thereby gives too much ‘weight’ to the extreme items in the stimulus-configuration we will not use the algorithm.  In this paper we will use PROXSCAL  that Busing proposed at SOFTSTAT in 1997 and that now has been implemented in SPSS as the counterpart of ALSCAL.  This algorithm minimises raw normalised stress.  The result is a far more honest Euclidean space.

This study investigates the psychometric properties, factor structure and validity of the revised Negative Acts  Auestionnaire-Revised (NAQ-R), an instrument designed to measure exposure to bullying in the workplace. By reanalyzing data based on a heterogeneous sample of 5288 UK employees, the results show that the 22-item instrument has a high internal stability,with three underlying factors: personal bullying, work-related bullying and physically intimidating
forms of bullying, although the instrument may also be used as a single factor measure. Criterion validity was explored by relating the scores on the NAQ-R to a single-item measure of perceived victimization from bullying, showing high correlations with both the total NAQ-R and scores on the three factors. Targets of bullying scored significantly higher on
all 22 items compared to non-targets. The NAQ-R correlated as expected with measures of mental health, psychosocial work environment and leadership, indicating a good construct
validity of the instrument. Furthermore, a latent class cluster (LCC) analysis showed that the instrument may be used to differentiate between groups of employees with different levels of exposure to bullying, ranging from infrequent exposure to incivility at work to severe victimization from bullying and harassment. The more commonly used operational criteria can also be used to detect targets of bullying. Hence, the NAQ-R is proposed as a standardized
and valid instrument for the measurement of workplace bullying.

This study investigates the prevalence of the four types of destructive leadership behaviour in the destructive and constructive leadership behaviour model, in a representative sample of the Norwegian workforce. The study employs two estimation methods: the operational classification method (OCM) and latent class cluster (LCC) analysis. The total prevalence of destructive leadership behaviour varied from 33.5% (OCM) to 61% (LCC), indicating that destructive leadership is not an anomaly. Destructive leadership comes in many shapes and forms, with passive forms prevailing over more active ones. The results showed that laissez-faire leadership behaviour was the most prevalent destructive leadership behaviour, followed by supportive–disloyal leadership and derailed leadership, while tyrannical leadership behaviour was the least prevalent destructive leadership behaviour. Furthermore, many leaders display constructive as well as destructive behaviours, indicating that leadership is not either constructive or destructive. The study contributes to a broader theoretical perspective
on what must be seen as typical behaviour among leaders.

This study investigates the prevalence of exposure to workplace bullying in a representative sample of 2539 Norwegian employees utilizing different measurement and estimation methods. The prevalence of workplace bullying varied from 2 to 14.3% depending on method of measurement and estimation.
Latent class cluster analysis is concluded to give the most reliable estimate (6.8%). Compared to a similar Norwegian study published in 1996, the prevalence of self-reported victimization from bullying is considerably reduced; from 8.6% in the early 1990s to 4.6% in 2005. In the 1996 study,4.6% labelled themselves as victims of severe bullying, the corresponding
number being 2% in 2005.