Huppert Check Fact Check – and the meaninglessness of parliamentary voting stats

April 16, 2015 § 10 Comments

So, recently I’ve been linked a few times by Labour activists to a website called HuppertCheck.info (http://huppertcheck.info) which claims the following “In deciding which party to back, voters will be interested in the record of the current Liberal Democrat MP, who is standing for re-election. Here we describe how Julian Huppert voted on key issues over the last five years.”

The site, as it fails to mention until you dig to the “about” page, is run by Labour activist Tom Yates, and is designed to attack Huppert’s record. Whilst it would be better if he’d actually put his affiliation more prominently, there’s nothing inherently wrong with that, and there are occasional times when the website makes entirely reasonable points.

But then there’s this graph, on the “An Independent Voice?” page. There’s a lot wrong with it, not least that percentages of votes are a meaningless statistic anyway (for reasons I’ll explain later). But that’s not really the major problem. The major problem is that the graph below, on Julian Huppert’s voting record, is false. It is a straight-up lie.

The above is simply untrue.

What they’ve done is taken the number of “rebellions” from the Public Whip website here. What they’ve totally ignored is the following statement, on the Public Whip site, about what their rebellion figures mean:

“Rebellion” on this website means a vote against the majority vote by members of the MP’s party. Unfortunately this will indicate that many members have rebelled in a free vote. Until precise data on when and how strongly each party has whipped is made available, there is no true way of identifying a “rebellion”. We know of no heuristics which can reliably detect free votes. See also the next question.

So actually, 912 is the number of times that Julian Huppert voted with a plurality of other Liberal MPs. That includes free votes – it also includes times when the Liberal backbenchers rebelled en masse, and times when the Liberals were whipped to vote against the Tories. As such, some examples of votes that are included as “supporting the tories” in HuppertCheck’s 912 figure:

  • Julian Huppert voting for the Affordable Homes Bill, which would have created bedroom tax exemptions anywhere where local councils were unable to provide alternative accommodation, and for many disabled people. All but one Tory voted against it.
  • Voting against Trident renewal in January 2015, opposed by Labour and the Tories.
  • Getting the equal marriage bill passed, in the face of Tory opposition.
  • A Lib Dem bill amendment that would have triggered by-elections when MPs were convicted of misconduct in public office. Both Labour and the Tories voted against the Lib Dems to block this.
  • Voting to block a Tory motion that would have promoted abstinence as a basis for sex education in schools
  • Julian voting on a free procedural vote on changing the times when the House of Commons sits on Tuesdays (which, once again, was opposed by more Tories than voted for it).

In other words, the graph does not show what it purports to, and includes a huge number of bills where the Liberals, or Huppert personally, voted against the Conservative party. The 912 vote, 97% figure given is a straight-up lie.It gets worse, though. Because this is not an honest mistake. Labour KNOW the above is a lie, because they have a different figure printed on their leaflets. A figure which puts Huppert’s rate of voting against the Tories up from three percent to voting against the Conservatives in nearly one in five votes. The Labour activists promoting this website have been delivering leaflets saying one figure and promoting HuppertCheck telling a direct lie that contradicts it.

Edit and clarification 01/05/15: HuppertCheck have now amended their page to give a more accurate representation of the data. I also wish to clarify, in case anyone was confused by the above, that I do not think that Tom Yates himself was knowingly lying or knew that Labour were promoting a different figure printed on their leaflets; I do, however, stand by my statement above that other Labour activists who were delivering the leaflets almost certainly did.


So, part two, some points on why this figure means relatively little anyway.

  • Most bills are only passed on their third reading; voting for the first or second reading of a bill only puts it in for further scrutiny. As such, it would be possible to oppose pretty much a government’s entire legislative agenda and still vote alongside them over sixty percent of the time.
  • The government votes figure is inflated by the presence of amendments purely designed to derail a motion, or by motions which are just frankly silly. For example, the government has to vote every year to charge certain taxes. The opposition generally vote against these motions – Labour literally voted against charging corporation tax and income tax at all several times in this Parliament. These, and the flip-side of opposition day motions that do the reverse, push these statistics up and achieve pretty much nothing.
  • There are also times when all the parties vote the same way, so the figures in the graph above include a number of motions that Labour voted in favour of.
  • And of course the Lib Dems did get to write some of the legislative agenda. The voting system referendum alone took thirty divisions to sort out, all for a measure the Tories ultimately opposed.
  • There’s also the fact that MPs don’t turn up for all votes (Julian’s figure at 77% attended is pretty high), and they’ll usually be more likely to appear for ones they think are important. Liberal MPs are, unsurprisingly, going to go through and vote on amendments (which is one thing that really boosts the figures here) far more carefully on Lib Dem sponsored legislation, whereas they might de facto oppose a whipped Tory-sponsored vote simply by not bothering to turn up.
  • Divisions aren’t everything Parliament does. Petitioning ministers, doing local casework, sitting on select committees to write reports on legislation, and things like Early Day Motions are all other parts of an MPs work.
  • What this means is that what’s important is less how many times you rebel, but where and when you do so. Julian has, for example, rebelled on the important third reading of the Health and Social Care Act over worries that it would lead to too much NHS privatisation, to vote against capping welfare increases below the rate of inflation, and to oppose secret courts, among other things.

Some conclusions: I’m not going to pitch this as a “therefore you should vote X way”. All I’m going to ask is two important things of you, dear reader.

  1. Please for goodness’ sake do not share HuppertCheck’s website until they have removed the above graph, and preferably until someone’s apologised for knowingly lying about this. Regardless of your political preference, it’s dirty, disreputable politics. Edit 01/05/15: As of now, HuppertCheck have amended their figures. Whilst I still consider their numerical voting percentage stats to be a very poor indicator of an MP’s independence or lack thereof for the reasons given above, their new numbers are, at least, more accurate. 
  2. Vote on the basis of what an MP supports, and don’t trust party partisan stats unless you really know what they mean and what the source data is. If you think that Julian voting to scrap EMA, to support Osborne’s budgets, and to initially introduce the Bedroom Tax are unforgivable, then you might want to consider voting against him. On the other hand, if you think his opposition to Tuition Fees, secret courts and NHS privatisation, his efforts to ameliorate the Bedroom Tax, and his advocacy for trans issues, cycling, and other things that get little airing in Parliament outweigh that, you might want to support him for re-election.

But don’t decide on the basis of meaningless statistics – especially when they turn out to be a direct lie.

Where Am I?

You are currently viewing the archives for April, 2015 at Thoughts of Progress.