As unlikely as it may seem, a market research survey caused a modest media storm this week. Football club Tottenham Hotspur sent an email survey to its fans in the United States. Among other things, they asked: “To what extent do you agree or disagree with the following statement; ‘A woman’s place is in the home’”. They provided the five standard response options from ‘Strongly Agree’ through to ‘Strongly Disagree’.

Some fans reacted negatively on Twitter, believing this question to be sexist. Spurs immediately capitulated. A club statement said:

“The survey questions were compiled by a third party on behalf of the club. The inclusion of this question in a club survey was wholly unacceptable and a regrettable oversight. It has been immediately removed from the survey for any other fans now looking to fill this out. We sincerely apologise to anyone offended by its initial inclusion.”

Forgive the pun, but that is terrible defending by Spurs! Trajectory was not the third party involved in this research, but we routinely ask the offending question in our global tracking surveys and would argue for the benefits in doing so. We also ask, “To what extent do you agree or disagree that when jobs are scarce, employers should give a priority to nationals over immigrants” (again with five response options from ‘Strongly Agree’ to ‘Strongly Disagree’). Does that make us xenophobic or anti-immigrant?

If so, we are in good company because we use these questions as part of our collaborative agreement with the World Values Survey, a global network of social scientists studying changing values and their impact on social and political life, led by an international team of scholars.

These questions have been asked in global surveys going back to 1981 and the results have been used in scores of peer reviewed academic papers, as well as more commercial work such as ours.

Our defence of these questions would be three pronged:

  • To ask for levels of agreement or disagreement with a statement that is inherently prejudiced does not make those asking the question prejudiced. To think it does is literally illogical
  • Understanding more about the profile of those with prejudiced views is a key instrument of policy to tackle prejudice
  • More specifically, there is real value in tracking these opinions over time. Individual, snapshot surveys can often under-estimate the true levels of prejudice. People can be reluctant to admit to it in surveys (though this is less of an issue in online research these days where there is no interviewer to offend). Longitudinal and/or trends research can give us a reading on prejudice’s direction of travel. If it is getting worse, resources and policy can be marshalled to tackle it

Clearly, this is sensitive stuff. Some questions used in tracking studies can come to look outdated. If we were starting tracking attitudes to gender in the workplace today, a statement along the lines of ‘women should have equal rights to any job and equal rights in the workplace’ may be preferable to the one in the Spurs survey. However, even now there are research benefits in putting both prejudiced and non-prejudiced statements in a survey and looking at gender prejudice from every angle. For example, I wonder if TGI still ask people whether they agree or disagree that “A real man can drink ten pints in a night”? They certainly used to.

Another factor in this story was surely that the question was being asked by a football club. Football has a reputation for male boorishness. Perhaps those complaining on Twitter, seeing that the survey came from a football club, assumed the worst (a prejudice of sorts, of course)?

The answer rests in what Spurs would have done with the data had they collected it? If they intended to use it to pander to the sexist views of those agreeing with the statement, they would clearly be in the wrong. If they were gathering the data to understand more about the social outlook of their supporters, they have absolutely nothing to apologise for.