Ipsos MORI are still working on the review of their polling methods they launched after deciding their London polls had indeed over-estimated Labour support in the mayoral race. Avid watchers of polls will probably have spotted that their political monitor never surfaced last month – I understand they carried it out, but aren’t publishing any figures until the review is completed.

In the meantime Mike Smithson has the lowdown on the first adjustment MORI are planning on making – having reviewed their samples MORI have found they have been over-representing public sector workers.

As regular readers know, for reasons that no one knows for sure – though it’s probably something as simple as how busy people’s lifestyles are or a simply attitudinal difference towards answering the door or phone to a stranger and giving up 20 minutes to answer impertinent questions for no reward – raw samples tend to over-represent Labour voters.

ICM and Populus get round this by weighting samples according to how respondents voted last time, which we call “past vote” or “political weighting”. While on paper this is a simple task (look up the 2005 election result, weight the sample to match. Finished), in practice it isn’t, because it has been proven that people do not accurately recall how they voted at the last election. When deciding what figures to weight past vote to, ICM and Populus therefore have to guess what level of “false recall” to factor in.

MORI have always recoiled from this form of weighting because of the “false recall” problem – they worry that the level of false recall could change suddenly in response to events, and weighting by past vote could end up dampening down genuine changes of opinion. The problem is that without some form of political weighting the samples remain biased to Labour, the standard demographic weights on age, gender, tenure and so on don’t correlate strongly enough with voting intention to cancel it out.

If their samples were over-representing public sector workers, then this may be a solution. It’s an area where people should be able to give accurate answers, the proportion of people working for the public sector doesn’t change vastly from month to month, it should be possible to get relatively reliable targets to weight to and – most importantly – it is a factor that appears to correlate with voting intention, so correcting it would serve to reduce or eliminate the Labour bias.

We await the full results of MORI’s review to see what effect it has. It will also be interesting to see if this skew towards public sector workers in present in other companies samples. It is far from a given that quota sampled face-to-face samples will have the same biases as quasi-random phone samples – they are obtained in different ways, but it is at least worth a look.


14 Responses to “Too many public sector workers?”

  1. For one crazy moment I thought oyur headline was a political question rather than a polling question. Which would make the answer, yes.

    On the polling question is this still a case of being different for the sake of it?

  2. MS poses this question :-

    “The question this raises is quite why a disproportionate number of public sectors workers answer the phone and agree to take part in interviews when the pollster calls. Is there something in the nature of the people employed or their working patterns that causes this to happen?”

    …mmmm, interesting question.

  3. It isn’t necessarily a case of more public workers being in the sample so the poll is biased towards Labour – the causal direction might be the reverse -this is sometimes in academic circles called an ‘ecological fallacy’:
    It could be that the sample is ‘too Labour’ for other reasons, and therefore as a result will include more public sector workers, as they are more inclined to Labour.
    Availability and accessibility are the main problems, and there has at times been a cultural differential refusal rate (though not, I suspect, in the current political climate).

  4. “The question this raises is quite why a disproportionate number of public sectors workers answer the phone and agree to take part in interviews when the pollster calls. Is there something in the nature of the people employed or their working patterns that causes this to happen?”

    I suppose it reflects how overworked and stressed they are!

  5. It seems more likely to me that the “man” in the street or the “person” at home during the day to answer questions face to face or on the phone is most likely to be unemployed and a definate Labour voter.

  6. Mike – no, that can’t be the reason. Work status (i.e. whether someone is employed full time, part time, unemployed, retired, etc) is something that is already weighted for (and in MORI’s quota samples, controlled in the quotas to). Samples do not over-represent the unemployed.

  7. “Sickness absence rates in the public sector and the private sector widened to record levels in 2007-Sickness absence rates in public sector outstrip private sector by 55% a major CBI/AXA report has revealed-”

    http://www.personneltoday.com/articles/2008/05/14/45837/sickness-absence-rates-in-public-sector-outstrip-private-sector-by-55-in.html

    I don’t know if this factor would have a “polling” effect?

  8. What percentage of the electorate are unemployed anyway – about 1 million out of 45 million – just over 2%? And all indicators are that they are less likely to vote than average, so often not included in polls.
    It has however been true that historically it is easier to find Labour voters. When I was a pollster with Harris, we always found that Labour were overrepresented, though that was in the days of face to face polls, when interviewers tended to go to densely populated areas, easier to fulfil their quota. There was also then the phenomenon of the ‘shy Tories’, though that probably is not true at the moment. I guess more middle class people have blockages to unsolicited telephone calls, may have more external demands on their time, and so on, though there are controls on class in all polls.
    All in all, polls are certainly much better linked to important variables now, so to assume Labour are overstated is less true than it used to be.

  9. I would say the next government should be able to severly cut the mumber of public sector white- collars.

  10. Public sector workers may be diverging somewhat now – perhaps those on the front line who generally have good reason to believe their practical job is important and not properly rewarded – even under Labour, and are more prepared to try another government,
    versus the massive public sector middle class of quangos, consultants, HR and so on who would fear a Tory government, and are a much better target for cuts, because if such people went, hardly anyone else would care.

  11. Public sector workers not hit by the 10p tax fiasco and long-term recipients of benefits are probably the only two categories of voters left in the country keen to retain a Labour Government at the next election.

    On the other hand many fully-employed,relatively affluent, cash-rich time-poor workers use their landlines as a call screening/answerphone service and rarely, if ever, pick up to an anonymous number. These will screen themselves out of polls. Many Conservative voters object violently to any invasion of their privacy (illustrated by the huge support for David Davis which has taken pollsters by surprise.) But they are more likely than the long-term unemployed to bother to vote.

    Labour’s share of the vote will continue to plummet regardless of poll results.

  12. I think the health and safety and outreach and awareness and enrichment department have plenty of time to answer the questions of pollsters. As a former public sector worker, I certainyl would have welcomed some entertainment during my 2 hour lunchbreaks when I had nothing to do.

  13. In response to Robert Waller, unemployment in this country isn’t one million but actually nearer four million. In 2007 there was around 900,000 unemployed and 2.8 million on sickness benefit. If you want to add in single-parent benefit increases of 350,000 from 500,000 to 850,000 then this increases the ecomically inactive persons living off benefits even higher. This actually contadicts the Governments false assertions that unemployment is at an all time low of one million.

    Unemployment was actually falling since around 1995/6 and the Labour Government inherited this trend. It continued to fall until 2005, due to both economic growth and more importantly around 600,000 extra jobs created in the public sector by this Government.

  14. Unemployment hovered at only just under 3 million in February/March 1993 – I expected it to go on rising at least for some months or a year, because after the 1980-81 recession, it went on rising until 1986.

    But in fact, it started to fall quite suddenly from about March/April 1993, and gathered pace quite quickly.

    I think the Labour market age had increased a lot in the 1980s, worsening the unemployment problem, because about 3 million net new jobs were created between 1981-86, but no fall in unemployment.