The NetMums survey: how not to conduct a survey

I’ve been trying to ignore this, as frankly it’s utter bollocks but people keep sending it my way anyway.

Transparent knock-off of Mumsnet, who are creatively called Netmums, have produced survey data declaring feminism is dead. Predictably, the media have leapt on it and are masturbating frenziedly to the whole thing. Now, obviously feminism isn’t dead, unless there’s some sort of Halloween witchcraft keeping all the zombie discussions of the liberation of women from patriarchy going, so that’s a non-point in the first place.

Now, there are discussions to be had about the label of “feminist” and whether or not people want to wear it, but this definitely shouldn’t be hung off of the Netmums survey, for the simple fact that the Netmums survey doesn’t actually show us anything whatsoever, except how to conduct one a survey really poorly and then get it into all the papers, thus promoting your brand as better than those splitters at Mumsnet.

(for what it’s worth, I proudly wear the label feminist, but see why others are reticent because there’s some utter wankers who also call themselves feminists storming around being thoroughly wrong about everything)

To conduct a good survey, you need a good sample of people to conduct the survey on. You want people who are representative of the population: a spread of races, ages, class and so on, in a way that is roughly similar to how things are distributed in the real world. It’s also best to seek out people to participate in the survey rather than stick it online and see who volunteers to take part–this helps make your data more representative. This is why opinion polling services are often engaged to gauge public opinion.

Netmums had a pretty big response to their survey–1300 people took part. However, only Netmums members were surveyed. They literally have a “surveys” section of their website, where users can participate in surveys to win prizes. This is hardly a particularly representative sample of women in the UK, and therefore it comes as no surprise that the survey data revealed that “women” want the next struggle for social liberation to be “reinstating the value of motherhood”. If you ask these questions only to women on an internet forum centred around the identity of “mother”, of course they’re going to say this is one of the most pressing issues to them.

The next issue in developing a good survey is how the questions are asked. Ask a leading question, and you’ll get the answers you want. Rather unsurprisingly, Netmums didn’t bother linking to a full set of survey questions, so it’s pretty much impossible to discern what they were actually asking participants and the order in which the questions were asked.

The glimpses we get suggest that the questions weren’t particularly well thought-out, asking to what extent people agreed with various statements like “I would like a bit of old-fashioned chivalry”. There’s a particularly risible question where participants are asked “Which of these activities are acceptable for feminism”, listing a range of options which respondents presumably tick, including highlights, baking cupcakes and prostitution. For real. Firstly, “acceptable for feminism” is a pretty complicated and confusing way of phrasing the question, which would mean different things to different people. Secondly, these things are all lumped in together to guide responses: it’s really no wonder that so many “women” think that sex work is unacceptable, when they’re comparing it to wearing false nails.

So it’s a pretty shittily done survey, and you’ll be delighted to know their reporting of it isn’t any better. Please study the image below for an example of how horribly they fucked it up (description below, for those using screen-readers):

Note that the text next to the pie chart–of presumably the same questions–bears literally no relation to what’s outlined in the pie chart. Admittedly, it’s a crap pie-chart and the weird oblong shape makes it very difficult to discern percentages, but absolutely nowhere is anything that can be remotely construed as anyone agreeing with a statement about feminism saying it’s not about equality, and the percentages don’t seem to add up at all. I don’t know if there were more questions asked or if Netmums are literally pulling things out of their arse, but either way it’s a thoroughly disingenuous way of reporting data.

Also noteworthy is what they don’t draw attention to in the blurb next to the pie chart: the fact that more than half of the respondents have said they don’t identify as feminists because they don’t need the label.

It’s a smart tactic, though. They stick the shit they want people to see on the press release, highlight the bits they really want people to see, and hope nobody actually bothers looking at their data. And they usually don’t, because journalists are lazy and they will therefore just spew out whatever Netmums told them to say.

Ultimately, then, let’s forget this survey ever happened. It’s a crap survey which tells us nothing. I imagine it’ll rear its ugly head periodically, so consider this a public service explaining why this is something to which no attention whatsoever should be paid.

_

Image depicts text and a pie chart. The text reads:

<“If you don’t call yourself a ‘Feminist’ why is this?”

39% criticised old-fashion Feminism for being too divisive, claiming they ‘dont want to be equal – women are different to men and we should celebrate the differences.’

Almost a third (28%) think traditional radical Feminism is ‘too aggressive’ towards men while a quarter (24%) no longer view it as a positive label for women. One in five describe Feminism as ‘old fashioned’ and simply ‘not relevant’ to their generation.

In subsequent questions, 17% even claim Feminism has gone too far, oppressing men and ‘losing sight of the natural roles of men and women’.>

The pie chart consists of five slices. Moving clockwise from the top, the largest segment, comprising more than half is coloured blue and labelled “I can be strong without labelling myself. A beige segment equivalent to roughly 10% is labelled “I don’t feel it’s relevant any more”. A turquoise segment, also around 10% is labelled “It’s old-fashioned, it’s not really my generation”. An orange segment, also 10% is labelled ” It’s not a positive label any more”. A pink segment, slightly larger than 10%, is labelled “It’s a bit aggressive towards men whereas we need compromise”.