Use and abuse of statistics II - Page 2

graeme · 2003-08-07, 13:26

Hah - Wackerbarth (what *is* that name anyway) catches me out again.

I did know that they do set small controled fires, of course. I don't suppose you'll let me get away with "that's not what I meant" in a puzzle forum (no - I didn't think so)

G.

asdf · 2003-08-07, 15:14

Fahrenheit 451 was a good book, and I thought of that as a good answer to that first one, but firemen is a vague term. Now it has been replaced by firefighter. The correct word could be "arsonist" or something like that.

graeme · 2003-09-03, 10:07

I'm half sorry to post this in the puzzles section, but it's almost an exact example of the use and abuse of statistics problem I posted above.

In the free paper today on the tube is this Huge Headline

Growing up into Junkies with the by-line of
A third of children have been exposed to hard drugs a disturbing new study reveals today

reading the article deflates all these claims of course. All my points above apply
there were 2,318 10-13 year olds in the survey, so not a particularly large study, but it seems that it was "country-wide" (whatever that means, it turns out that the survey was done in TWO areas - Glasgow and Newcastle).

They gave the kids a questionaire, and we know that everyone tells the truth on questionaires.

It looks like 4% of 12 year olds have used drugs - but reading a bit more indicates that this included alcohol. (and frankly I'd be surprised if a 12 year old hadn't had a glass of wine or a shandy)

6% of 11 year olds had used drugs in the last year again I guess this also includes alcohol.

It turns out that "exposed to drugs" means something like "seen someone take some" but again this is unclear.

1% had tried heroin - or at least said that they had. It turns out that this is 1% of users, so this is 1% of 4% of 2,300 - hmm less than 1 person had used heroin (or said that they had anyway).

I hope the real document is better than this report of it, because it seems like another waste of time and effort producing inconclusive results, about a serious subject.

If Professor Neil McKeganey is reading this, please, please, tell me that I'm wrong and this is a properly done study, so that any decisions made on it will actually have some solid foundation.

2003-09-03, 10:07	#14
graeme Jul 2003 41 Posts	In the news I'm half sorry to post this in the puzzles section, but it's almost an exact example of the use and abuse of statistics problem I posted above. In the free paper today on the tube is this Huge Headline Growing up into Junkies with the by-line of A third of children have been exposed to hard drugs a disturbing new study reveals today reading the article deflates all these claims of course. All my points above apply there were 2,318 10-13 year olds in the survey, so not a particularly large study, but it seems that it was "country-wide" (whatever that means, it turns out that the survey was done in TWO areas - Glasgow and Newcastle). They gave the kids a questionaire, and we know that everyone tells the truth on questionaires. It looks like 4% of 12 year olds have used drugs - but reading a bit more indicates that this included alcohol. (and frankly I'd be surprised if a 12 year old hadn't had a glass of wine or a shandy) 6% of 11 year olds had used drugs in the last year again I guess this also includes alcohol. It turns out that "exposed to drugs" means something like "seen someone take some" but again this is unclear. 1% had tried heroin - or at least said that they had. It turns out that this is 1% of users, so this is 1% of 4% of 2,300 - hmm less than 1 person had used heroin (or said that they had anyway). I hope the real document is better than this report of it, because it seems like another waste of time and effort producing inconclusive results, about a serious subject. If Professor Neil McKeganey is reading this, please, please, tell me that I'm wrong and this is a properly done study, so that any decisions made on it will actually have some solid foundation.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Statistics	mahnouman	Information & Answers	4	2013-02-13 13:31
You know what they say about statistics ...	petrw1	PrimeNet	1	2007-10-08 13:29
but this is abuse.	R.D. Silverman	Soap Box	34	2006-01-10 17:51
Statistics	R.D. Silverman	NFSNET Discussion	1	2004-06-14 18:40
Use and abuse of Statistics I	graeme	Puzzles	1	2003-08-07 12:49

2003-08-07, 13:26	#12
graeme Jul 2003 29₁₆ Posts	Hah - Wackerbarth (what is that name anyway) catches me out again. I did know that they do set small controled fires, of course. I don't suppose you'll let me get away with "that's not what I meant" in a puzzle forum (no - I didn't think so) G.

2003-08-07, 15:14	#13
asdf Sep 2002 2²×3×5 Posts	Fahrenheit 451 was a good book, and I thought of that as a good answer to that first one, but firemen is a vague term. Now it has been replaced by firefighter. The correct word could be "arsonist" or something like that.