Welcome Guest! The IOSH forums are a free resource to both members and non-members. Login or register to use them

Postings made by forum users are personal opinions. IOSH is not responsible for the content or accuracy of any of the information contained in forum postings. Please carefully consider any advice you receive.

Notification

Icon
Error

Options
Go to last post Go to first unread
Admin  
#1 Posted : 06 June 2006 15:12:00(UTC)
Rank: Guest
Admin

Posted By Jonathan Breeze Well, I've finally convinced Management to issue the HSE Stress Survey (only taken 18 months) and come across a problem: Some of our employees (whom I personally suspect would have the most to gain by getting things off their chest) are now refusing to fill in the HSE Stress Survey on the grounds that they do not feel the questionnaire is sufficiently anonymised and the data could be used to identify them. The questionnaire was approved by management, unions and the safety committee who all felt it was suitably anonymous, yet requested enough info to target action departmentally if needed. We even sent a covering letter explaining why we were doing this with links to the HSE website. My question is, how do I go about preventing self selecting responses occurring in the future?
Admin  
#2 Posted : 06 June 2006 16:49:00(UTC)
Rank: Guest
Admin

Posted By Jeffrey Watt JB I'm not up on the parlance. What do you mean by self selecting responses. Sorry my stoopid chip is in overdrive today. Jeff
Admin  
#3 Posted : 06 June 2006 17:04:00(UTC)
Rank: Guest
Admin

Posted By Jonathan Breeze Okay, it may not be the correct technical term, in which case I apologise. My theory is: The very people who are refusing to fill in the questionnaire are, I suspect, under stress in the workplace, hence their paranoia (though this may be correlation rather than cause & effect). This means that the responses I receive will generally favour those who are not currently suffering stress in the workplace and as a consequence the data will generate a false positive result. Is that any clearer?
Admin  
#4 Posted : 06 June 2006 17:28:00(UTC)
Rank: Guest
Admin

Posted By Kieran J Duignan For starters, Jonathan, it's a good idea to avoid language such as 'paranoia' - unless you really want to trigger unresolvable levels of conflict-riddent stress. Perhaps the core concepts are really 'defensiveness','anxiety'and - especially'-'a creative touch'? A 'competent' outsider - e.g. a chartered occupational psychologist who is also a chartered safety and health practitioner and a qualified counsellor - may have the advantage that he/she can be wheeled in as a 'neutral' competent person, listen and report and leave. He/she can avoid unfounded allegations of bias and contamination, to a degree that is rarely possible when an employee, however pure his motives, takes the lead. He/she can also be versatile, e.g. offer to conduct a focus group or use the comparatively powerful repertory grids, as an alternative to written survey. An independent competent person can also act in a shrewdly and creatively confronting style: I have, for example, on occasions informed staff that they forfeit their statutory rights to complain about an employer where they prevent me, as the appointed independent 'safety-competent person', from conducting a valid assessment. This normally results in high-quality co-operation; in the rare event where it didn't, I have fulfilled my 'threat' and gained reasonable compliance by principled confrontation (which most staff applauded very enthusiastically because they strongly resented the waves habitually created by self-centred bullies). Emotional and political tensions about stress are often similar to those that used to arise about redundancy management when that was still 'hot'. Yet for about a decade, I worked intensively in several organisations as a 'redundancy, career and stress counsellor', sometimes at the active request of 'aggressive' trade unions who were persuaded I had no hidden agenda (and became clients themselves when their turns to exit came along).
Admin  
#5 Posted : 07 June 2006 09:01:00(UTC)
Rank: Guest
Admin

Posted By Jeffrey Watt JB Got you now 100%. I think Kieran's post has validity re getting in an independant party to conduct stress audits. But it doesn't work 100%. Some people always believe in management collusion no matter how much contrary evidence you provide. JB you obviously have the best gut feel for your organisation but here is a suggestion. Perhaps a staged approach with the TU and a counselling service may act as a confidence building measure to tackle the percieved issue of anonymity. This could then pave the way for a more effective audit. Best of luck Jeff
Admin  
#6 Posted : 07 June 2006 10:47:00(UTC)
Rank: Guest
Admin

Posted By Jonathan Breeze Kieran, Fair point about my phrasiology there, I was intentionally being blunt to simplify the point and stimulate a discussion. I do like the alternative descriptions you have given, however I don't want to couch the terms too much either. All, Discussing the situation with a Data Analyst of my acquaintance, a nul response from specific departments might actually be a result we can investigate by itself, even if it does not fit in neatly with the HSE Indicator Tool. However it still doesn't solve the problem that I need to ask staff certain questions about e.g. gender, which department they are they in, and whether they are F/T or P/T in order to meaningfully analyse the results. If I do this I can't help but identify some people in the organisation (e.g. in our female biased organisation I am one of only two male head office staff). I'm also in the situation where baseline data is needed by the board before funds could be released to approach a 3rd party - but that's another issue.
Admin  
#7 Posted : 07 June 2006 11:08:00(UTC)
Rank: Guest
Admin

Posted By Tabs Maybe you should look at your analysis targets first? For instance, why do you want to know which department to look at for a particular problem? If one department has a certain problem, are you advocating providing 'correction' just in that department, and not horizontally throughout the company? (if so, how do you catch the areas which don't want to raise the issue, or the next area the indivduals move into? - or indeed the interaction between departments?) If I was providing, for instance, training to correct sexual harrassement issues, I would not target one department... would you? If you want or need to provide anonymity, you really should not be able to even guess whether the results you are looking at comes from your CEO or the person sat next to you. To do that though, means you have to flatten your responses as well as your questions. I would suggest that where people have raised this as an issue, they have a better insight into recrimination than you might be aware of - so you should seriously consider the cross-company responses.
Admin  
#8 Posted : 07 June 2006 11:46:00(UTC)
Rank: Guest
Admin

Posted By Jonathan Breeze That's actually quite helpful Tabs. We thought we had already done that to an extent during the last X months of discussion. But the gender question will always pose a problem in any organisation or department where there is a clear majority of one gender. So we would therefore need to remove that particular question in our organisation. However how would we then be able to find out whether workplace stress was more prevalent in a particular gender of employee?
Admin  
#9 Posted : 07 June 2006 12:14:00(UTC)
Rank: Guest
Admin

Posted By Jeffrey Watt JB I face the same problems. If it is not specific enough to identify individual departments because we may put employees anonymity at risk how can it be of use in helping individuals with specific problems, I would suggest it can't. I think any organisation under 400 people has a real difficulty using this tool in the type of way we as safety professionals want it to work i.e. identifying specific areas that require an intervention. I would suggest it is of better use as a self assessment tool linked to wider training and self referring support mechanisms. Discuss chums Jeff
Admin  
#10 Posted : 07 June 2006 12:39:00(UTC)
Rank: Guest
Admin

Posted By J Knight How we get around the departmental identifier is we ask the respondents to put the answers into labelled envelopes; the question sheets themselves cannot identify a respondents location. The location batches are processed one location at time. I am the only person who ever opens the envelopes; I keep all the completed quesrionnaires in a locked drawer; I am not in line of management to anybody in the Charity except for my Retail H&S Adviser & he's not stressed ;-). We have only had problems with fears about anonymity in one workplace, and it only affected two returns, John
Admin  
#11 Posted : 07 June 2006 14:12:00(UTC)
Rank: Guest
Admin

Posted By Jeffrey Watt John I am already bearing those kind of measures in mind. A 3rd party organisation is doing the lot, posting the forms, analysis, focus groups. People will still have fears regards identification and self select as to Jonathon's original question in his post. All I'm trying to say is that if you take the measures that you and I mention then you are doing enough. That will not be good enough for some people but I think it is good enough from discharging your duty of care at this stage of the process. Do you concur/disagree? Jeff
Admin  
#12 Posted : 07 June 2006 14:34:00(UTC)
Rank: Guest
Admin

Posted By Jonathan Breeze Jeff, That was my original proposal in our organisation and for much the same reasons. However the counter argument I put forward (because I believe that management should see the full picture) was that the HSE had published this survey & analysis tool and that we could hardly be censored for using the system that they had created. In the end they went with the HSE Tool administered internally.
Admin  
#13 Posted : 07 June 2006 14:37:00(UTC)
Rank: Guest
Admin

Posted By J Knight Jeff, I agree, we did the same with our Organisation Climate Survey which was an HR initiative. Jonathon, We use the straightforward HSE question sets as well; though we have opted for the first draft rather than the final version, John
Admin  
#14 Posted : 07 June 2006 14:45:00(UTC)
Rank: Guest
Admin

Posted By garyh I suggest that you either get the results sent directly by the respondents to an independent body or to a trusted on site "friendly face" and give an absolute guarantee of anonymity. Managers just get to see the anonymised results. Having done this survey in large and small organisations, at very different levels of safety excellence, the results were similar. I suspect that it gets skewed by "whingeing" however it is a great starting point for worker involvement and consultation.........not the end point. Doing something with the results is the hard bit.
Admin  
#15 Posted : 07 June 2006 15:17:00(UTC)
Rank: Guest
Admin

Posted By Max Bancroft If you get a 50% response rate, do you assume that the 50% who didn't respond are not stressed at all? Or that they are stressed out due to the bad management and don't want to respond because they think nothing will happen due to the bad management? Or they think they can be identified? Or a mixture of all three and if so, in what proprtions?
Admin  
#16 Posted : 07 June 2006 15:33:00(UTC)
Rank: Guest
Admin

Posted By Jonathan Breeze Max, We don't know because they haven't been sampled. However, when you hear conversations along the lines of "I'm not going to respond because they'll be able to work out it's me from my answers." Then it does raise alarm bells that those with either negative experiences (or possibly a negative outlook) are not being sampled. What it 'could' mean is that you cannot assume that the responses you have received are in fact random, but that the methodology of using questionnaires to measure stress in the workplace is flawed in the first place.
Admin  
#17 Posted : 07 June 2006 15:53:00(UTC)
Rank: Guest
Admin

Posted By Max Bancroft Thanks Jonathan - I was hoping that some others who are more expert in the field of stress and stress surveys would be able to say. Of course, the answer may be different in each organisation. We ran a stress survey. We got a 45% response and I've been wondering ever since - is eveything OK out there or is there an unexploded bomb! Of course, the survey is only part of managing stress.
Admin  
#18 Posted : 07 June 2006 16:00:00(UTC)
Rank: Guest
Admin

Posted By J Knight Jonathon, I think you probably have it right; there are flaws in the method, and it is a problem if there is an atmosphere of recrmination (real or imagined). Significantly, the one workplace where we definitely had people afiling to respond because of fear is one where we know that the staff culture is one of conflict. In this case, the Manager knows what is happening, she just needs the stress survey as a way of taking heer concerns back to the workforce for discussion; a way of opening things out, if you like, John
Admin  
#19 Posted : 08 June 2006 10:24:00(UTC)
Rank: Guest
Admin

Posted By Tabs Jonathon, sorry I'm puzzled by your response to my post - why do you need to know if one gender is more stressed than another? The stated aim of stress management is to reduce harmful stress in the workplace. I can't see how gender is relevant to that aim, unless you want to identify people (hence their scepticism?). I know different people have different stress raisers, but these are what should be explored during consultation with individuals. I thought your initial questionnaire is to get a base measurement. At this stage I would much rather submit a report that says "Two people find working for their present manager stressful" than to submit "Two women in the packaging area..etc." You may even find that by making the report as blind as possible would be to remove potential prejudices in senior management.
Admin  
#20 Posted : 08 June 2006 10:53:00(UTC)
Rank: Guest
Admin

Posted By Jonathan Breeze Tabs, Your raising of that very question in your initial response was helpful, in that it alerted me to the fact that the question may not be necessary. Some members of the reviewing panel had previous experience of preparing equal opps questionnaires, which may have clouded the issue. Also there was an assumption that this data would be useful in identifying if pregnant employees were at particular risk. Thus the potential consequences of exposure to stress were perhaps being confused with whether individuals were suffering from stress in the first place. Does that clarify what I meant? I'm frustrated now that the whole thing was approved by so many people without anyone picking up on it at the time.
Admin  
#21 Posted : 08 June 2006 11:03:00(UTC)
Rank: Guest
Admin

Posted By Kieran J Duignan In so far as 'stress' is a professional problem - and much work-related stress is not! - accurate, well-informed diagnosis is critically important. As Max suggests, survey is only one method of assessment. Often it is far from appropriate; where it is, staff have the legal right to oppose it. It is up to the 'safety-and-health competent' person to avail of the many, many, many more appropriate methods of assessment available to deal with the issues 'as far as reasonably practicable'. Regrettably, Max's use of the expression 'unexploded bomb' - like Jonathan's earlier choice of 'paranoia'- illustrates inappropriate, confusing and potentially counter-productive use of language about a complex phenomenon. Is it not the responsibility of safety professionals need to avoid the use of stress-inducing labelling apt to make matters worse?
Admin  
#22 Posted : 08 June 2006 11:41:00(UTC)
Rank: Guest
Admin

Posted By Diane Thomason Kieran Do you mean language such as "self-centred bullies"?? (as in your post above)
Admin  
#23 Posted : 08 June 2006 16:28:00(UTC)
Rank: Guest
Admin

Posted By Maggie Atterbury Hi Jonathon I would actually stop trying to please everyone. If the Safety Committee (with employee representatives) and the Trade Union Reps are satisfied with the questionnaire, you should go ahead and issue it. Don't try and guess why individuals do not complete the form, as you will not get 100% returns with the best will in the world. Everyone will have been given the chance to give their views and the important bit is what happens next. If it becomes clear that where definate concerns were raised, there were no recriminations, but positive action resulted, everyone will be much more likely to take part next time. This is what happened with us.
Admin  
#24 Posted : 08 June 2006 17:23:00(UTC)
Rank: Guest
Admin

Posted By H Klinkenberg Jonathan As a freelance professional corporate stress manager I have worked on many stress risk assessments and stress audits with many organisations of all sizes. I was involved in stress auditing long before the HSE introduced their management toolkit. The best I have achieved where completing a voluntary completion questionnaire on its first run was 80%, the usual percentage is between 40% and 50%. As the questionnaire should be repeated in about 18 to 24 months, getting a better percentage on the second run will indicate an increased trust in management. Thus you have shown a measurable improvement with regards to ‘trust in management’, which used to be a significant stress indicator in the old Stress Audits. You are trying to find stress issues, not prove they don’t exist. Management would like to believe that they don’t exist but we all know the truth. Getting to opinion of over 40% of the workforce is a significant result that gives you a great starting point for introducing stress risk minimisation. Regards Humphrey Klinkenberg Freelance Stress Management Consultant and Trainer
Users browsing this topic
Guest (4)
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.