This week I'll be blogging about The Economist's "Anglo-Saxon" attitudes survey. You can find the survey itself here. I was particularly struck by the throw-away comment:
[Americans and Britons] feel much the same about the death penalty: they are broadly against it. - [Source]
I thought this was surprising, as the survey appears to show nothing of the sort. From the survey itself (Question 4, page 9 of the pdf file):Do you favour the death penalty for murder? (Yes, always; Yes, sometimes, depending on the circumstances; No, the death penalty is wrong; Don't know)
| Britain | USA | |
| Yes, always | 21 | 26 |
| Yes, sometimes | 53 | 50 |
| No | 24 | 20 |
| Don't knows | 2 | 5* |
* NB: yes, the American data does add up to 101%: suspect either a rounding error or typo.
UPDATE: The Economist retracts; see next post.
5 comments:
The errors on that survey must be pretty large - and splitting by political affiliation will end up nigh on meaningless.
Not to mention the myriad other limitations of survey data.
But at least they do the very minimum necessary to take a survey seriously - i.e. provide full questions and answers, and details of the sample. Didn't see a response rate though - but that is widely seen as the Achilles heel of survey data (making, for instance, the published conifidence intervals meaningless).
"But at least they do the very minimum necessary to take a survey seriously - i.e. provide full questions and answers, and details of the sample."
It certainly makes a nice change from the Metro/Daily Mail style pluck-stat-from-out-of-thin-air-and include-in-quirky-paragraph-to-fill in-the-gaps-between-the-adds.
Re:response rate, I think YouGov (who did the UK end of the survey)
use telephone banks, so the death penalty question alone shouldn't skew the survey.
I am always deeply skeptical about "values" surveys like this, as I think the response is strongly susceptible to what happened in the last week: some particularly egregious miscarriage of justice, or some equally horrific murder?
However, the issue I took with survey was that they make a statement that is simply unjustified by the results.
I was going to blog about the religion questions today, which are so badly structured as to render the data completely meaningless. Instead, I wrote a long and spectacularly tedious chunk of thesis. Because that's the exciting kind of life I lead.
Hopefully, I'll attack them later this week.
...and The Economist retracts:
In our article on Anglo-Saxon attitudes (”Anglo-Saxon attitudes”, March 29th) we said that Americans and Britons felt much the same about the death penalty: they were broadly against it. They do feel strikingly similarly, but not as we said. Between a quarter and a fifth are opposed, the same proportion are in favour, and around half would support the death penalty in some circumstances. Apologies. The online version of the article has been corrected.
Heh, I imagine such questions as:
"Which do you think is more likely to be the correct explanation for the origin of the earth?
The theory of evolution
The account of creation as told in the Bible
“Intelligent design” – evolution has happened but by intelligent design
Not sure"
Will be attracting your attention.
"Heh, I imagine such questions as:
"Which do you think is more likely to be the correct explanation for the origin of the earth?
The theory of evolution
The account of creation as told in the Bible
“Intelligent design” – evolution has happened but by intelligent design
Not sure"
Will be attracting your attention."
Oh yes, indeedy. :)
Post a Comment