It’s not as bad as you think
The Royal Statistical Society has just commissioned an opinion poll to look at beliefs about policy-relevant issues and how they relate to reality. The first few results:
- Teenage pregnancy: on average, we think teenage pregnancy is 25 times higher than official estimates: we think that 15% of girls under16 get pregnant each year, when official figures suggest it is around 0.6%.
- Crime: 58% do not believe that crime is falling, when the Crime Survey for England and Wales shows that incidents of crime were 19% lower in 2012 than in 2006/07 and 53% lower than in 1995. 51% think violent crime is rising, when it has fallen from almost 2.5 million incidents in 2006/07 to under 2 million in 2012.
- Job-seekers allowance: 29% of people think we spend more on JSA than pensions, when in fact we spend 15 times more on pensions (£4.9bn vs £74.2bn).
- Benefit fraud: people estimate that 34 times more benefit money is claimed fraudulently than official estimates: the public think that £24 out of every £100 spent on benefits is claimed fraudulently, compared with official estimates of £0.70 per £100.
You might look up what the actual figures are in New Zealand. To get you started, Paula Bennett claimed about $200 million in benefit fraud in 2010/11 (and about 10% of that was prosecuted) from about $9 billion, or about 22c per $100.
Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »
The bit about teenage pregnancy was a bit disingenuous – and I am surprised that you didn’t make a QA comment. Teenage pregnancy goes until someone runs out of the XXteen bit of their lives.
In New Zealand the 15-19 birth rate (so ignoring abortions) is 0.02724 but the rate for 15 year olds is 0.00413 and 14 year olds is 0.00087.
Ignoring girls under 14 this suggests that the teenage pregnancy rate as a whole is almost 11 times the rate for 14 and 15 year olds — which assuming the UK is similar to here — means people overestimate the number but the difference is between an actual 6.5% and estimated 15%.
Mary Smith
11 years ago
They make the correct comparison for the question they posed in the poll. They asked, “in your opinion, what proportion of girls under the age of 16 years in Britain get pregnant each year?”
11 years ago
The question asked the poll was about pregnancy before age 16: all the questions and the full data are linked from the RSS post. The question was “In your opinion, what proportion of girls under the age of 16 years in Britain get pregnant each year?”
The lowest answer given was 1%, by 5% of respondents. The mean was 15%. The truth is 0.6% (ignoring girls under 13).
If the actual pregnancy rate up to and including age 19 is still less than half what people thought it for ages 13-15, that makes the finding more dramatic, not less.
You mention abortions. The data includes both pregnancies resulting in birth and abortions. That’s also detailed in the documents linked from the RSS post. According to this report (table 4a) there are about 3 abortions per 1000 in this age range, so that’s about half the pregnancies.
11 years ago
Nice exercises, I’ll try one :)
From http://www.stats.govt.nz/tools_and_services/nzdotstat/recorded-crime-statistics/ASOC-offence-calendar-year-statistics.aspx and using offenses per 10000 pop,
~1300 for all offenses in 1996, and has declined more of less steadily (R^2 of 0.87 on the annual decline of 1.6%) til last year it was only ~850.
Colbert’s quip that “reality has a well-known liberal bias” is half right.
If however, you believe that the public are exercised by problems completely out of scale with their severity, and that this provides cover and incentive for political parties to pursue agendas contrary to the public weal… then it’s just as bad as you think :(.
I note that one of the study’s recommendations is to educate a public so that this kind of thing is well known. As it’s my belief that this is what education is supposed to be doing already, how likely is this to work? Don’t we have (my own prejudice here possibly) one of the most literate, numerate, populations that there have ever been?
11 years ago
I was musing about this yesterday. Or rather about the why the disconnect exists between what is and what is believed. Especially given that the data appears to be freely available.
The only possible cause that I could come up with was the emphasis put on these things by the media. i.e. continual focus on the horrific crimes leading to the perception that violent crime is rising etc.
Any other/better explanations exist that I’m not aware of?
11 years ago
It’s not just the media: we all have the same biases to some degree.
We focus more on rare or dramatic events than routine ones, we care more about not losing money than gaining money, and we particularly focus on being ripped off. These are well-documented cognitive biases, and it’s easy to see that (especially in a village society) they are useful biases to have.
The Daily Mail is just universal human cognitive biases writ large.
11 years ago
I’m probably overestimating the medias ability to recognise and compensate for those biases. It would be grand if quality media that claimed to be all about bringing the truth to the public could do so (
I first saw these stats reported in the guardian)
Describing the mail as human cognitive biases writ large though, that’s a gem :)
11 years ago