Posts by David Hood
Last ←Newer Page 1 2 3 4 5 Older→ First
-
I accidentally left the camera on high ISO after taking pictures in poor light, went out into the sun, and took this.
-
With some hurried lunchtime calculations. Taking only the general election election results (so assuming current polling, and the actions of the party, doesn't actually matter), and abusing a loess line into something of a straight regression so I can extend it, we set the centre of the range of likely results for the 2014 election as being 32.35% of the vote for Labour, 42.59% for National.
So, as a starting point (and acknowledging that this is ignoring polling), I'd be curious as to what would be the effect with 32.35% of the vote (I realise it will be fairly close to the figure for 33.5% based on current polling earlier, so one or two more men).
csv of election results
https://www.dropbox.com/s/vhw3mhgga02npac/LabNatVot.csv
hastily written R code
https://www.dropbox.com/s/6xut1xo6wi2h8ve/LabNatVot.R -
Speaker: Gender quotas (and helping…, in reply to Graeme Edgeler,
just used the result pages and a calculator.
Good enough, thanks for the links
-
Speaker: Gender quotas (and helping…, in reply to Graeme Edgeler,
Don't make me test this assumption by going over past lists!
but will it be a statistically significant number of demotions compared to prior to the policy. Since there are more men, the raw number of men likely to be demoted is likely to be higher than the raw number of women. But to numerically make the cause that the policy caused (or even the numerically softer correlates to) demotions, that would have to shift the proportions before and after the policy, the standard amount of change is just noise for the purposes of this question.
-
Speaker: Gender quotas (and helping…, in reply to Graeme Edgeler,
we can look to history
Graeme, do you have a quick source to the time series you used for this? I'd like to play along at home (or at least during my lunch hour) and am hoping not to have to spend the time to assemble it myself.
-
It seems appropriate to link to the new free online data journalism course on next year
http://datadrivenjournalism.net/news_and_analysis/Data_Journalism_Handbook_Creators_Announce_Online_Data_Journalism_Course
(as seen on StatsChat) -
Speaker: Gender quotas (and helping…, in reply to Andrew C,
You cannot fancy pants the maths to get around this.
So, your objection is that, while no current male MPs will suffer (the news story in the question) new MPs (and both sexes will be equally qualified to be there) are statistically more likely to reflect what the general population is like?
-
mostly don’t watch at all. They are incredibly creative and communal, too. I believe they are what is known as “the future”.
I showed my daughter this comment and she went "That's me"
-
Sun through tree
-
Shag in Leith, Univeristy of Otago
