Do women make better constituency MPs than men? That, anyway, was the claim made during one of this week’s many Parliament Week activities. It’s a claim you’ll occasionally hear – although it is much less common to hear any evidence to support it.
As it happens, one of us is currently ploughing his way through a dataset of public attitudes towards MPs, which contains several questions from a survey asked in July this year that would allow us to test this. Here is a quick and dirty check of this claim.
The most obvious test would be respondent’s satisfaction with their local MP (‘Are you satisfied or dissatisfied with the way your local MP is doing their job?’). There is a difference here. Of respondents who had women MPs, some 34% said they were satisfied (that is, either Very Satisfied or Fairly Satisfied combined) with their local MP; some 35% said they were dissatisfied (Fairly Dissatisfied or Very Dissatisfied). That’s a net score – those satisfied minus those dissatisfied – of -1. Those with male MPs scored 29% satisfied, 37% dissatisfied, a net score of -8. This difference, however, is not statistically significant.
The survey included other three measures that might detect a gendered constituency focus. The first asked whether the MP was seen as more locally focussed or nationally focussed (or somewhere in between)? There was no difference between male and female MPs at all, both seen as predominantly locally focussed by 53% of respondents. The second asked whether the respondent had contacted their MP in ‘the last two or three years’. Being more approachable might be one indicator of being a ‘better’ constituency MP. In total, around 21% of respondents claimed to have contacted their MP in the last ‘two or three’ years. But the figures for those with male MPs (21%) and women (23%) were virtually identical. And then third, if they had approached their MP, were they satisfied with the response? Here we have to be careful, because when you take a normal-sized sample, examine the minority of respondents who have contacted their MP, and then split that sample yet again into those with male MPs and those with female MPs, the N involved is fairly small (c.350), particular in the case of those with female MPs (c.70). But for the record, some 55% of those with male MPs were satisfied with the response they got from their MP compared to 60% of those with women MPs, another statistically insignificant difference.
In other words, it’s difficult to detect any obvious signs here of women making better constituency MPs. There are some differences but none that achieve statistical significance.
First, as with all surveys, it’s always possible that there is some nuance in behaviour that is not being detected by the questions asked in the survey. Maybe. But if women do make ‘better’ constituency MPs, they do so in a way that does not lead to their constituents being more satisfied with them; that does not lead to their constituents thinking they are more focussed on the constituency; that does not lead to them being more likely to be approached and that does not lead to them dealing with those constituents who do approach them in a way that makes them more satisfied. For sure, these aren’t the only measures of ‘success’, but they are hardly insignificant either.
Second, as noted above, with a normal sized sample, it is difficult to test some of these claims as well as one would like. Women MPs are still only a minority of the house, and so in any normal sample only a minority of respondents have a woman as their local MP.
And third, this is, as noted above, a pretty quick and dirty check of this theory. To do it properly – to control for the various other factors that might be driving (or masking) any differences would require some multivariate analysis. But initial attempts – this piece is being written before breakfast… – didn’t yield anything.
Based on this, then, it is seems difficult to make the claim that women are better constituency MPs than men. Nor, indeed, are men better constituency MPs than women. Perhaps that’s a better outcome?