Bartholomew's Notes

Unity at Ministry of Truth has the remarkable story of 10Yetis, a PR firm which regularly announces the results of surveys on behalf of clients, with the hope of getting the clients’ names in the media:

Looking down the list of press releases, I feel that we should be eternally grateful to this company for the fascinating sociological they provide into the nature of modern British society. Without their sterling efforts, how else would we find out that half of all women believe they perform better at work when they dress ‘sexily’, or that the most accident prone people in Britain live in Dundee, let alone that the most pampered women in Britain are in Birmingham while Essex girls are apparently the cheapest dates – mmm, I have my doubts about that one.

Of course, complaining about tabloids using this kind of thing as the basis for some cheap and easy copy is probably futile, but one can reasonably expect a bit better from the BBC:

More than half of parents do not think sex education should be taught to children in school from a young age, a survey suggests.

Of 1,700 parents of UK 5-11 year olds surveyed by the BabyChild website, 59% said they disagreed with the practice.

The most common reason given was that it is “inappropriate to teach children about sex”.

The survey comes after a Bill calling for girls to be “taught to say no” passed its first reading in parliament.

In this case, “BabyChild” isn’t a client of 10Yetis – it’s actually another business run by Andy Barr, who co-owns 10Yetis (Barr also provided a soundbite for the Daily Mail on What It All Means). The Bill  – brought forward by Nadine Dorries MP – was discussed in Parliament a few days ago, and the survey provided the hook for Babychild to get some free publicity on the coat-tails of the subsequent media discussion.

Tim Ireland asked to see the actual survey at the weekend, and was sent a two-page document which breaks down the results for each question. However, there is no methodology, and the number of “respondents” is equal to the number the BBC says was “surveyed” – in other words, non-respondents are not factored in, which suggests the sample was self-selected from the internet (most likely via the site SurveyMonkey). Unity explains:

…their sample methodology is best described as ‘anyone who can be arsed to fill in an online poll’, which I think neatly explains why the company isn’t a member of the Market Research Society and isn’t, therefore, required to muck around with of the awkward stuff, such as using representative samples and compiling demographic data of survey respondents, that other, reputable pollsters have to undertake.

The survey results are, to say the least, short on demographic information. We’re told only that 1732 people completed the survey, all of whom are parents of children aged between 5 and 11, and that last fact alone explains why the survey came down so strongly against sex education in schools.

The first question the survey asks is ‘Do you agree with the fact that sex education is often taught to children in schools, even from a young age?’, leaving respondents to figure out for themselves what ‘sex education’ actually means.

The Market Research Society (or MRS) is, according to its website

the world’s largest association serving all those with professional equity in provision or use of market, social and opinion research, and in business intelligence, market analysis, customer insight and consultancy.

…As ‘the voice of market research’, MRS defends and promotes research in its advocacy and representational efforts.

…Through its media relations and public affairs activities, MRS aims to create the widest possible understanding of the process and value of market, social and opinion research, and to achieve the most favourable climate of opinion and legislative environment for research.

It seems to me that determining whether a private company is a member of the MRS should be first point of enquiry for any journalist reporting on polls and surveys from such a source.

Filed under: Uncategorized