**Mx56 :**

Wee factual quibble with Marsh’s bit. 80% of Leave supporters did not say that immigration is a force for ill in that Ashcroft poll. 80% of people who thought that immigration was a force for ill in that poll voted Leave.

If you look at the data tables for the poll (which are on Lord Ashcroft’s website), on page 168 you’ll see a table based on ratings from 0 (very much a force for ill) to 10 (very much a force for good), Ashcroft seems to have taken 0-4 as “Force for ill” and 6-10 as “Force for good” with 5 as “Mixed blessing”. The resulting contingency table should look something like this (knocked together myself in Excel because Ashcroft is a git and presents his data in a hideously unfriendly format) – http://imgur.com/lMB2Wse

From that we have 62% of Leave voters thinking immigration is a force for ill vs. 17% of Remain voters. It’s not that the point you were making was wrong, just that the stats were off and that’s kind of important when we’re talking about people being insufficiently informed.

(and for anybody who cares, the chi square test statistic from that would be a whopping 3269.80, for a p value of infinitesimally tiny size)

Thanks very much, the correction is much appreciated!

]]>I can already see you are going to get a lot of angry post and debate over the doughnut issue. ]]>

If you look at the data tables for the poll (which are on Lord Ashcroft’s website), on page 168 you’ll see a table based on ratings from 0 (very much a force for ill) to 10 (very much a force for good), Ashcroft seems to have taken 0-4 as “Force for ill” and 6-10 as “Force for good” with 5 as “Mixed blessing”. The resulting contingency table should look something like this (knocked together myself in Excel because Ashcroft is a git and presents his data in a hideously unfriendly format) – http://imgur.com/lMB2Wse

From that we have 62% of Leave voters thinking immigration is a force for ill vs. 17% of Remain voters. It’s not that the point you were making was wrong, just that the stats were off and that’s kind of important when we’re talking about people being insufficiently informed.

(and for anybody who cares, the chi square test statistic from that would be a whopping 3269.80, for a p value of infinitesimally tiny size)

]]>And Mike is right, doughnuts are not *specifically* a breakfast. #teammike

The latter half of this podcast was a nice distraction from the country being fucked.

]]>