The 2019 intake of Conservative MPs into UK's parliament numbered 365. Since then at least 6 Tory MPs have been suspended or resigned for engaging in sexual related misconduct. These are:

  1. Julian Knight - accusation of sexual assualt
  2. Rob Roberts - sexting
  3. David Warburton - sexual harassment
  4. Christopher Pincher - groping scandal
  5. Imran Khan - sexual assault
  6. Neil Parish - viewing pornography in the house of commons

That means that roughly 1.6% of the parliamentary party has either resigned or been suspended due to sexual misconduct which seems to be a high figure. To put into perspective, the UK working population in December 2023 was roughly 31 million. A similar rate of misconduct amongst the general working population would see nearly 500,000 workers resign or be removed in the 5 year window starting from December '23.

So my question is, is this rate of sexual misconduct actually exceptional when compared to the general working population, other parliamentary parties in the UK, and other elected houses in similar socia-economic countries?

  • In case someone is confused en.wikipedia.org/wiki/Imran_Ahmad_Khan (Not my DV though.) Feb 19 at 23:36
  • 4
    re. downvote and discredit... I can't read the OP's mind so I can't vouch for things - but will assume their good intent in asking this question. Thing is, before you assume the OP just wants to drag poor Tories in the dirt, rest assured that many opposition politicians will try to "make hay" from these events. Asking here whether these events are indeed exceptional in frequency and hopefully getting some perspective in answers seems a very valid political question. I agree with ohw's answer, but if an answer justifiably argued the opposite, that would also be informative. Feb 20 at 3:05
  • 1
    @ItalianPhilosophers4Monica, I can assure you, or anyone else concerned, that I have no political axe to grind here. Just saw the article and was surprised at the numbers.
    – Ben Cohen
    Feb 20 at 7:11

1 Answer 1


is this rate of sexual misconduct actually exceptional when compared to the general working population, other parliamentary parties in the UK, and other elected houses in similar socia-economic countries?


The rate at which elected officials are discovered to have engaged in misconduct tends to be higher than average, in part, because these officials are under constant high scrutiny compared to the average person (so they are more likely to be caught), and in part, because power hungry people who sometimes abuse their authority or are dishonest, are disproportionately attracted to pursuing careers in politics.

But, this number of sex offenses would still be exceptional, even for elected officials, in either the U.K. or elsewhere in the developed world, although not wildly so.

This is about 0.4% per year, which would be equivalent to two offenses in one two year Congressional session from members of a single party. This is not something that has never happened (the most common issue in the U.S. Congress is inappropriate behavior towards teenage "pages" on the staff of Congress that comes up now and then). For there to be four such offenses in the U.S. Congress in a four year period, however, would be notable, but probably not entirely unprecedented.

Compared to the rate of "all causes" serious misconduct, this is less exceptional. If there were no other forms of serious misconduct in the session (perhaps defined as conduct giving rise to an expulsion or resignation), and this time around all of the serious cases just happened to be sex offenses as opposed to, for example, financial corruption or domestic violence or substance abuse related cases or other non-sexual violent conduct, that would make the overall figure less surprising.

Also, not all of these matters are equally serious. Neil Parish's sad but victimless offense does not compare to the far more serious charges against Imran Khan. The number looks less exceptional if only the most serious offenses are included.

The U.K. Has A Political Culture With High Standards Of Conduct

The U.K., due to the strong role of political parties in keeping members in line and its general political culture, secures resignations in circumstances involving even comparatively less serious indiscretions or accusations, where politicians in other countries might be censored or fined, but not forced to resign, and might be more prone to dispute the charges formally.

For example, the Lauren Boebert Beetlejuice scandal (involving publicly vaping and public consensual sexual groping while attending a live musical performance in Denver, Colorado, with a man who wasn't her husband while she was still legally married), didn't result in any formal consequences or meaningful demands to resign in the U.S., but would probably have resulted in her resigning if at the time she had been a Tory MP, instead of a Republican member of Congress in the U.S.

Long Term Trends Are Probably Less Extreme

To some extent, however, this may also be a product of what is sometimes called the "law of small numbers" which is that some uncommon event may seem much more frequent than it is in the whole data set, when you are only looking at a small part of the data set and the absolute number of rare cases is small.

Even a single sexual misconduct allegation in a four year period, in this small population, so any event at all, looked at in a short enough time period, will far exceed the rate of this kind of conduct in the general public.

If you looked at rare episodes of misconduct among such as small group of people, over a much longer time period, it is likely that you would see long periods of time with no misconduct, and that is the kind of comparison you would need to make in order to make a meaningful comparisons to rates in the general public or in other groups of elected officials.

Put another way, small datasets from a larger complete sample are more prone to statistical flukes.

For example, to look at a much longer dataset (skewed due to charges not being brought in minor cases due to historical immunities for aristocrats):

There were 30 House of Lords convictions of peers resulting in a punishment for the peer (in 29 cases a death sentence, and in the one final case, in 3 months of imprisonment) from 1499 until the practice was abolished in 1948, a time period covering all but the first 158 years of the 607 years during which House of Lords trials were available and the process was formalized.

Of those convicted peers, 30% were pardoned or commuted by the monarch. . . . .

Incidentally, 30 convictions of offenses limited to murder and treason in this time period isn't vanishingly low. Before the recent contraction of the House of Lords, it had about 730 peers at any one time, so in the 449 year period for which we have records, there were about 328,000 person years of potential defendants, with a conviction for murder or treason every 11,000 or so person years. This is quite high by modern standard for an exceedingly privileged and carefully socialized group of people who didn't have access to firearms for most of that time period.

The Long Term Trend Is Towards Better Behavior

I would also suggest that there is a long term secular trend towards conduct improving in legislative bodies. They were a lot more rowdy in the 18th century than they are today, and lawlessness tends to be more common among elected officials in young democracies than in older ones.

  • 1
    Nice answer. I am doubtful whether Lauren Boebert's behaviour would have resulted in her resigning as an MP in the UK. Matt Hancock famously kissed his lover during COVID lockdown and did not resign as an MP (though he did have to step down as health minister). It would be good if you could add references for the claims regardign rates of sexual offences in other parliaments / elected bodies.
    – Ben Cohen
    Feb 20 at 7:10
  • On the law of small numbers, one can turn this into a classic statistics question. We assume that the annual rate of sexual misconduct cases in parliament is the same as in the overall population (null hypothesis) and then compute the chance to see 4 cases in 2 years given this base rate. This gives you a result of the form 'we can conclude with 99% confidence that the base rate in parliament is higher than in the population as a whole' combined with some estimate of what the base rate in parliament could be.
    – quarague
    Feb 20 at 8:05
  • I am also dubious whether pornography consumption is a victimless crime, but that is a discussion for elsewhere.
    – Ben Cohen
    Feb 20 at 11:07
  • @quarague "compute the chance to see 4 cases in 2 years given this base rate" Over 365 parliaments over many centuries? I haven't run the numbers (in part, because they are hard to compare). But looking at just two years in a small population and comparing it to the entire population of the U.K. isn't really a valid comparison.
    – ohwilleke
    Feb 20 at 15:04
  • Based on the overall population you compute some average rate for the number of sexual misconduct cases per year. One can then ask whether 4 cases over 2 years with 365 members of parliament is consistent with this average number and just some bad luck or whether that much bad luck is unlikely and the rate of cases in parliament is higher than the overall population rate. This is a statistics questions and there are well-defined meanings of 'bad luck' and 'how unlikely'.
    – quarague
    Feb 20 at 18:08

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .