This will hurt me...

I was playing with some numbers yesterday that came across my desk and fed into the Department of Health and Human Services State of Public Health 2008 report and was surprised by a little quirk that I really hadn’t thought about. Now, we probably are all intrinsically aware of the clear divergence in health status between the most disadvantaged and most advantaged groups in our society. Statistically, lower socioeconomic status (SES) is associated with higher disability levels and higher premature death rates. Low SES – whichever way you choose to measure it: income, educational attainment or occupation – means poorer health, a higher incidence of chronic conditions and higher levels of health care utilisation. Moreover, health inequalities do not just indicate disadvantage, they tend to drive further disadvantage. Now for those who do not know, Tasmania – for a lot of historical, geographical, economic and sociological reasons – trails the rest of the country (bar the Northern Terri...

The non-profit, non-partisan Statistical Assessment Service (STATS) is an excellent resource on the use (and abuse) of science and statistics in the media. An affiliate of George Mason University in Virginia, STATS sets about identifying and correcting scientific misinformation in the media resulting from bad science, politics, or a simple lack of information or knowledge; and to act as a resource for journalists and policy makers on major scientific issues and controversies. Their website canvases a wide variety of issues. Some of their discussions that interests me most include What Science Really Says About the Benefits of Breast-Feeding ; The Risks of Television ; the links between autism and vaccines ; and an excellent and balanced exploration of the much commented upon Lancet study into deaths in Iraq, The Science of Counting the Dead . Now if it sounds like a promotional blog here, it sort of is. I'm pilfering something I've written for work-related purposes promoti...

I was set the challenge to find some statistical evidence of NSW bias in the selection of Australian Test cricketers. I can’t seem to find a state-by-state breakdown of Test players, and couldn’t be arsed going through every one, so I’ve had to find a different way to establish it. I decided to have a look at state players who have amassed significant figures at the first class level, but did not get much of a run at the higher level, and see how players from NSW compared with players from elsewhere. I’m not claiming that each of these players SHOULD have got a greater run in tests, merely comparing how much of a go high run-scorers and wicket-takers have received in test cricket. Personally, I found the results surprising, and believe that they tend to support some sort of institutional bias towards NSW representatives. Even considering the factor of being the largest state with the most cricketers, I think that the statistical breakdown establishes a pretty convincing argument that p...

I worked this out in early December, and have decided to post it here now because it's all very sparse. Brett Lee’s test bowling: 1–10 tests: 930 runs for 48 wickets @ 19.37 11–20: 1166 r 32 w @ 36.43 21–30: 1283 r 32 w @ 40.09 31–40: 1481 r 42 w @ 35.26 41–50: 1342 r 42 w @ 31.95 51–55: 583 r 17 w @ 34.29 His first 15 tests Lee had conceded 1523 runs for 65 wickets @ 23.43 . Since then, he has conceded 5262 runs for 148 wickets @ 35.55 . For a total of 6785 runs for 213 wickets @ 31.85 . As a comparison: AJ Bichel: 1870 r, 58 w @ 32.24 . DW Fleming: 1942 r, 75 w @ 25.89 . JN Gillespie: 6770 r, 259 w @ 26.13 . (First 35 tests @ 26.24 ; Second 36 tests @ 26.03 ). MS Kasprowicz: 3716 r, 113 w @ 32.88 . GD McGrath: 11787 r, 549 w @ 21.46 . PR Reiffel: 2804 r, 104 w @ 26.96 . Other bowlers and their averages who’ve played around the BLee era (albeit with far fewer games/opportunities): J Angel, 46.30 ; NW Bracken, 42.08 ; SR Clark, 17.39 ; SH Cook, 20.28 ; AC Dale, 31.16 ; BP Julian, ...