Reality bites

Reality bites - Research has found that job satisfaction scores differ depending on whether the questionnaire was completed by employees on a Monday or a Friday.

by RICHARD REEVES, a consultant and business writer
Last Updated: 31 Aug 2010

Research has found that job satisfaction scores differ depending on whether the questionnaire was completed by employees on a Monday or a Friday.

Please circle the statement that most closely reflects your own view: i) I love my mum; ii) I like my mum; iii) I neither like nor dislike my mum; iv) I dislike my mum; v) I hate my mum. And now the following: i) My favourite activity is filling out surveys; ii) Surveys are a painful necessity; iii) If I am asked to fill out another sodding survey, I will personally travel to the office of the sender and suffocate them with their own questionnaire.

Surveys are now part of everyday business life. Most of you probably have at least one uncompleted survey lurking in your inbox about which you feel vaguely guilty but for which you never quite have the time. We now have a small industry in surveys. Never before in the field of human endeavour has so much been asked of so many by so few.

But the blizzard of questionnaires yields little of real value and in many cases produces bad data, on which bad decisions are subsequently made. For a start, many of the questions are terrible. Some almost guarantee a certain answer, the social acceptable one - even if the truth is more complex. Social scientists call these 'yea-saying' questions. Examples include 'I love my mum' and 'Are you prepared to pay more taxes in order to fund better public services?' (a yea-saying question that destroyed the pollsters' predictions for the 1992 general election).

Commercial examples include: 'Is knowledge-sharing important for competitiveness?' and 'Is a diverse workforce good for business?'

More specific questions are much better. For example, a uselessly high percentage of people agree with the statement 'Marriage is a good thing'. More interesting data would be derived from: 'Is your marriage a good thing?'

It is vital therefore to treat most survey results with a large bucket of salt. At best, they give an impressionistic sense of people's views. It is not science. Research by Dr Mark Taylor of the Institute for Social and Economic Research at the University of Essex has found that job satisfaction scores differ depending on whether the questionnaire was completed on a Friday or a Monday.

Employee surveys also have a nasty habit of becoming self-fulfilling prophecies. If a staff survey finds that people think 'morale is low' - another classic example of a non-specific question - and the results are published, pretty soon everyone starts shaking their head about how low morale is and, before you know it, everyone has talked themselves into a greater state of misery than before.

None of which is to say that companies should not ask their employees for their views and their feelings. But the questions need to be designed to generate real data, the results should be disseminated with care and the whole exercise treated with the necessary scepticism.

Otherwise, there's a real danger that the whole employee survey exercise descends into corporate narcissism, and that companies spend too much time gazing and picking at their navel, rather than getting on with the job at hand.

Job satisfaction is a good case in point. Some people are perfectly satisfied to be comfortably paid to sit around doing as little as possible except annoy their colleagues. Is their 'satisfaction' good news?

On top of the internal surveys, there are, of course, a few forests' worth of external polls. In part, this is driven by the media's love of a survey. Journalists need a peg for a story, and a survey-derived fact usually does the job. 'Four out of five new mums have less sex than before'; 'One in four 20-somethings will remain childless'; 'Two-thirds of firms struggling to fill skills gaps' and so on. Few journalists care about the quality of the data. But most of these findings would be laughed at by the editor of an academic journal, who would ask questions about the survey response rate, validity of the sampling method and neutrality of the question phrasing.

Surveys that produce lists are even more enticing - so much so that there is now some competition to produce a list of the best companies to work for in the UK. The established Best Companies list, published by the Sunday Times, is going head-to-head with a new survey, the Best Workplaces in the UK, to be printed by the Financial Times. The methodology of Best Companies Ltd - for whom, to declare an interest, I have done some work - is sound; I can't vouch for the other candidate. (In both cases, of course, firms should be sure to get their staff to complete the questionnaires on a Friday.)

It is simply absurd to have two national surveys trying to do exactly the same thing, wasting time and resources all over the place. The DTI, in typically decisive New Labour style, is backing both rather than taking a stand.

At this rate, a survey of the time spent filling out surveys may soon be necessary. And then perhaps a survey to generate a list of the best surveys of the best companies.

This survey madness has to stop. But don't take my opinion. Let's ask everyone.

Find this article useful?

Get more great articles like this in your inbox every lunchtime

Subscribe

Get your essential reading delivered. Subscribe to Management Today