Evaluating stress measurement questionnaires
3 January 2002 -
* Do stress measurement questionnaires really measure stress in a reliable and valid way?
* Do they actually provide the information that organisations need to tackle workplace stress?
In a paper presented at the British Psychological Society Occupational Psychology Conference today researchers are revealing the findings from the first ever large scale review of stress measures. The strengths and weaknesses of different approaches to measuring stress are highlighted in their report which also discusses implications for organisations trying to measure and tackle stress at work.
The report was commissioned by the Health & Safety Executive (HSE) with a remit to review measures of workplace stressors - measures of those aspects or characteristics of jobs, such as workload or lack of control, which, when present at excessive levels, are believed to lead to poor psychological or physical health.
A team of independent organisational psychologists led by Dr Jo Rick from the Institute for Employment Studies (IES) and Dr Rob Briner from Birkbeck College, University of London condcuted the research.
They used a set of rigorous standards to evaluate over 25 different stress measures which are widely used in the UK. The researchers came up with a number of surprising findings:
1. The amount and quality of evidence they could find about different measures quite limited. In fact there was only sufficient evidence to allow a detailed analysis of 5 measures. The lack of evidence suggests that many stress measures have not been adequately developed. In many cases it seems that we do not know if these instruments are accurately measuring stress at all!
2. Even where particular stress measures are supported by evidence, results are inconsistent and mixed in a number of ways. This suggests that these stress measures are not very reliable tools for assessing workplace stress.
3. Most surprisingly, there was an almost complete absence of evidence about the predictive power of these stress measures. This a worrying finding because the main reason for measuring stress is to assess aspects of work which are likely to lead to health so that these harmful aspects can be changed. But it seems that virtually all the available evidence comes from one-off 'snap-shot' studies and these cannot show if the stressful aspects of work tapped by these measures actually lead to ill health.
Dr Jo Rick (Principal Research Fellow, Institute for Employment Studies) said:
'I was very surprised by the lack of evidence linking the workplace stress measured by these scales to possible ill-health outcomes. This has serious implications for organisations using these measures help them tackle stress at work.'
Dr Rob Briner (Senior Lecturer in Organizational Psychology, Birkbeck College) added: 'This report shows the need for a fundamental rethink of the way in which stress is measured at work and how more valid and reliable tools for assessing stress can be developed.'
What are the implications of the research? Organisations are required to assess stress but this research indicates that the tools by which they can make an assessment have severe limitations. In fact, organisations which are using commonly available stress measures may not be accurately measuring aspects of the work environment that might lead to ill-health. So they could be focusing on relatively harmless issues and missing real stress problems in the workplace.
The researchers argue that 'it is not clear, therefore, that using these measures either on their own or as part of a broader stress assessment process fulfils the requirements of health and safety legislation to identify and control those aspects of work that are likely to lead to ill health or harm.'
They conclude that we need more information about the reliability and validity of existing stress measures but it may be that the existing approach to stress measurement is questionable and new methods and techniques should be considered.