I was thinking the other day, I am totally immune to any argument based on the use of statistics. I just tune out. I believe that basically anyone can gather statistics to 'prove' their case, but I have no desire to check they are accurate, read up on the methods used to gain these statistics, think about the wording of the questions if a survey was involved, the motives of the researcher, where their funding has come from, the accuracy of the interpretation of the data, look into other similar studies etc etc. Then after all that you still often have to explain the difference between correlation and causation to the person quoting the stats.

Having a life, and many interests and hence topics I would have to read up on, I just don't have the time or energy to look into the accuracy of all these statistics bandied about as fact every day. It seems from some of the posts on OLO, some people have many hours to scroll through various documents and web sites and do all this sort of thing. I suspect though, that the average person...

a) Just either takes statistics and reported research by the media at face value. Which poses a real danger of people ignoring their own common sense, and lazily accepting that anyone who has these mythical numbers to back them up should be listened to.

b) Just doesn't believe any statistics, which makes all these studies useless as a tool for informing the community.

Ironically, I actually have a degree with a major in statistics:-) Not that I have ever used my once held knowledge for anything but understanding the futility of gambling.

Dear Usual Suspect,

The purpose of statistics is not just important to statisticians, or scientists, it's important to everyone.

But don't take my word for it.

See the following website:

http://www.kevinbone.com/statisticsfor.html

Statistics are just a form of data. The same applies with data as a whole as it does with stats; the problems of what importance you place on what, what’s causal, what’s independently derived, etc.

If a data set is anything other than very simple or very strongly indicative of a certain outcome, it is wide open to differences in interpretation, and hence the conclusions reached.

I’m often amazed within my professional botanical arena at some of the data that is presented and the conclusions drawn. This concerns things such as the basis for describing new plant species or for splitting off new genera, or for justifying changes from remnant to non remnant vegetation.

It's not rocket science, just descriptions and/or measurements on a relatively small number of characters/parameters, and yet, I can often see things very differently to the conclusions presented.

Particularly concerning the constant stream of scientific name changes of plants as new species are described and new genera circumscribed, I often have to say; you’ve got to be kidding! Other times I’m doubtful, but it would take a lot of research to confirm or remove the doubts. And often, I can agree.

It’s all over the place. So in the wider world, where there are many other forces at play; politics, profit motive, funding, etc, you do indeed have to be very careful about much of the data presented and the conclusions drawn.

.
Foxy, I can’t get that link to work. Nor the www.kevinbone.com homepage.

Dear Ludwig,

My apologies. I made a typo. The website should read:

www.kevinboone.com/statisticsfor.html

The title is:

The K-Zone: What is Statistics actually for?

Damm lies by statistics

give me your mubers ;i will rate them down to numbers per 100,000 [making them look smaller
then use simple accounting modifiers to tell at a glance what you need them to be saying

Take booze it kills 5000 per year , but take out violent alcohol related crime,road deaths you get it down to a basic number that becomes 2000 that further modified becomes MINUS 2000[simply subtract a statistical bennifitiary [ie bennifit derived from having one drink per day ]

the number becomes more reasonable [read the figures yourself google up legal drug deaths
[where the biggest 'cause' of death becomes smoking [how many died from the act of smoking is irrelivant because death is deemed [blame-able to any given cause cause
we dont autopsy deaths under docter supervision ,their cause of death is taken at face value

noting furyther 1 in 100 hospitalisations is from adverse drug reaction to legal perscribed medication
[one in ten causing death, any nicoteen stained finger [or ''attributable '' to smoking] becomes death by smoking ?

please see this joke about the latest scam
the NNT number

[doses Number Needed to Treat
[ie cure one person]using the neo[new drugs]

docters-dirty-little?-secret.

http://articles.mercola.com/sites/articles/archive/2008/05/08/medicine-s-dirty-little-secret.aspx?source=nl

NOting hemp-marijuanna [which cures cancer ] has never caused any death ,but cant be patented.

a New York City public school Last year, announced that students would only need the correct answer on 27 percent of the questions, 23 out of 84, on the Math A Regents exam in order to receive a passing grade.

In 2003, the score required for passing was a not-exactly-lofty 55 percent. But after two-thirds of the students failed that year, the State decided that rather than focus on fixing the students, it was better to just dumb-down the test.


Statisticlly Things do not appear as dismal in the City’s elementary schools.

Last year, Mayor Michael Bloomberg and City School’s Chancellor Joel Klein announced that fourth-graders had aced the State’s standard math exam, with 77.4 percent of all students meeting or exceeding State standards—the highest number for a City class in the history of the test. “said Bloomberg almost one year ago. “More fourth grade students are meeting or exceeding standards than ever since we started standards-based testing, and significant gains by Black and Hispanic students show that we continue to close the achievement gap as well.
The mayor failed to mention his initiative to end social promotion the year before, by which he held back the lowest performing students in third grade only, ensuring an increase in test scores that could be announced during the stretch run of his reelection campaign.
extracted from
http://www.fireflysun.com/progressiveed.htm
more statistical deciets

I'm at a bit of a loss to know what you're getting at with your opening post, Usual Suspect.

Have statistics somehow let you down somewhere?

Or is the winter of your discontent a season brought on by somewhat careless use of the word 'statistics' by others to encompass recorded measurements or counts? I know I could be accused of such careless use of the word.

I frequently refer, for example, to 'electoral statistics', when what I really mean to refer to are the actual records of votes counted, or electors enrolled. Work with such reams of mind-numbing records should perhaps better be described as 'internal checking' or 'auditing', rather than statistical analysis. Such work seldom requires more mathematical ability than the simple arithmetical operations of addition, subtraction, multiplication, or division. It is within the capability of almost all who can vote. Perhaps its very ease, and mind-numbingness, are reasons why few actually perform such cross-checks; such being, in a way, beneath their dignity. Hidden truths are thereby sometimes (frequently?) overlooked.

I guess its one of life's little coincidences that the title of this thread caught my attention because I had recently used almost the identical expression as the concluding line in a post to the topic 'Driving Down Our living Standards By Stealth', here: http://forum.onlineopinion.com.au/thread.asp?discussion=1725#33794 . Even more curiously, the post to which I was responding was one upon which Julie Vickers (of bra-snapper, chair-sniffer 'fame') saw fit to comment, five posts later, in a manner suggestive of its having been posted by a glove-puppet! And to think, Julie was engaging in a little political manipulation of her own with the rest of her post!

At least my conscience is clear. The internal checking I am wont to do with electoral 'statistics' shouldn't let anybody down, and thus could not have contributed in any way to your disillusionment with statistics.

It could, perhaps show some people up, though.

Have a nice day. Put your talents to good use.

Carthago delenda est!

The best misuse of statistics is the constantly regurgitated phrase about the average womens wage being less then the average man's wage. They fail to tell you that women engage in more part time work than men and that account for most of the difference

Actually Atman, such an obvious omission is actually corrected for. The truth of the statement remains.

So many people think that the statisticians at the ABS seem to be no better than high-school remdial maths graduates.

It must be remembered that statistics are not lies, they are reflections of actual data points. It's the interpretation and attribution of the underlying causes of trends that gets most people into trouble.

Bugsy,

' It's the interpretation and attribution of the underlying causes of trends that gets most people into trouble.'

Ha, too true. Though you wouldn't see anyone on OLO mis-using stats deliberately or not.

I would say though that probably the hours worked in the studies Atman is discussing would be capped at 'full time'. In many industries this is 60+ for high earners with no parental responsibilities or a partner at home full time, which is more likely to be a bloke.

US “Ironically, I actually have a degree with a major in statistics:-) Not that I have ever used my once held knowledge for anything but understanding the futility of gambling.”

Strange US, you are an objective fellow, I would have thought it came as second nature to you.

I studied Stats as part of my accountancy grind. I have used them often, as they support processes like CPA and PERT. Here I challenged Fred Argy on the merit of tax. (The data from nation master, if you compare “happiness” to “Tax as % of GDP” illustrates a significant negative correlation).

Stats is good stuff. Sure the results can be manipulated particularly with loaded questions and skewed sampling (like testing the general view on abortion outside a catholic or LDS church) but if you are confident of the validity of collection the output is the most objective analysis possible

Col,

I think I'm more averse to people who quote stats in their arguments, and media reports based on a 'new study'. As I said I tune out and find them totally unconvincing, but more because of the reasons I outlined re checking up on them.

With regards to politics, I find Yes Minister highly amusing...

Sir Humphrey Appleby: Mr. Woolley, are you worried about the rise in crime among teenagers?
Bernard Woolley: Yes.
Sir Humphrey Appleby: Do you think there is lack of discipline and vigorous training in our Comprehensive Schools?
Bernard Woolley: Yes.
Sir Humphrey Appleby: Do you think young people welcome some structure and leadership in their lives?
Bernard Woolley: Yes.
Sir Humphrey Appleby: Do they respond to a challenge?
Bernard Woolley: Yes.
Sir Humphrey Appleby: Might you be in favour of reintroducing National Service?
Bernard Woolley: Er, I might be.
Sir Humphrey Appleby: Yes or no?
Bernard Woolley: Yes.
Sir Humphrey Appleby: Of course, after all you've said you can't say no to that. On the other hand, the surveys can reach opposite conclusions.
[survey two]
Sir Humphrey Appleby: Mr. Woolley, are you worried about the danger of war?
Bernard Woolley: Yes.
Sir Humphrey Appleby: Are you unhappy about the growth of armaments?
Bernard Woolley: Yes.
Sir Humphrey Appleby: Do you think there's a danger in giving young people guns and teaching them how to kill?
Bernard Woolley: Yes.
Sir Humphrey Appleby: Do you think it's wrong to force people to take arms against their will?
Bernard Woolley: Yes.
Sir Humphrey Appleby: Would you oppose the reintroduction of conscription?
Bernard Woolley: Yes.
[does a double-take]
Sir Humphrey Appleby: There you are, Bernard. The perfectly balanced sample.

some statistics make sense--for example the IQ's of populations that have fluoridation have dropped by 4 to 6 points over a decade.
[US,Japan and China]

Australia the UK and US--presently tick off 86,000 [australian figures only] ADHT kids as sensible--

containing the same methedrine they use for combat troops if you wonder why the occassional classroom slaughter or teen killing engages our attention.

For those with better to do than Alberts mastery of figures--ie two explanations for the one phenomena of relativity--perhaps we should try modelling mathematically the real world:

"Mathematics, material science, is indispensable to the intelligent discussion of the material aspects of the universe, but such knowledge is not necessarily a part of the higher realization of truth or of the personal appreciation of spiritual realities.

Not only in the realms of life but even in the world of physical energy, the sum of two or more things is very often something more than, or something different from, the predictable additive consequences of such unions.

The entire science of mathematics, the whole domain of philosophy, the highest physics or chemistry, could not predict or know that the union of two gaseous hydrogen atoms with one gaseous oxygen atom would result in a new and qualitatively superadditive substance--liquid water.

The understanding knowledge of this one physiochemical phenomenon should have prevented the development of materialistic philosophy and mechanistic cosmology." [hubert wilkins,1934].

For those that believe statistics are valid--would suggest they
attempt to explain the NON-LINEAR reactions--which is exactly why fluoride, radiation and synthetic medications cause such havoc-- ie there is no safe dosage where harm doesnt occur.

Its not surprising then that if statistics may illustrate so much--that since 1934 you have not been able to PUBLISH honestly.

the linear folk are that devoid of talent...

US I like your example of “loaded questions” LOL.

And just love that show (Yes Minister / Prime Minister), one of the funniest ever produced.

I think statistics are like rectal probes:

They should only be used by someone who is proficient and for what they do, it is difficult to find an alternative.