Just saw an interesting statistic on TV..

A

AmadeusM

Banned
Joined
Jul 1, 2004
Messages
1,799
A story by Diane Sawyer on ABC about staying at home moms vs. career moms..

Anyway, the stat was that after divorce man's standard of living goes up 10%, and the woman's standard of living goes down 27%.

I am interested in what makes those 10% happen?

Thanks.
 
I would ask first what that 10% (and 27%) refers to. Disposable income? Improved career possiblities? Party time? (:))
 
Whenever one sees an "interesting statistic," one should ask, "Where did that number come from? Who measured it? How was it measured? What definitions were used? How was the data collected? How were the questions asked?" Remember 47% of statistics are just made up.

Just the other day, I heard an interview with a woman talking about a particular cause of death (no need to be specific here). She said, "Over a thousand teenagers died from this in the US last year!" (I was suspecious about that number immediately because we're only a matter of weeks into 2006 and it takes longer than that to compile valid statistics about 1000+ deaths across the US. Many of those deaths probably haven't even had an official determination of cause issued yet.) The reported picked up on that and was smart enough to ask, "Over a thousand deaths per year!?! Wow. Where did that number come from?" The response was, "Ah.... well... that's just our estimate. A lot of these deaths are mis-classified." Count that statistic among the 47%.
 
For example, I could ask 1000 randomly-selected Americans attending a Hillary Clinton Fund Raiser, "Do you think George Bush wears good ties?"

The answers might be, "I hate the man and everything he does.... but I do have to admit that he wears nice ties, so yes."

I then translate that and report that 87% of Hillary Clinton Supporters approve of Pres. Bush.

It's all in the wrist.
 
Lies, damn lies and statistics.
- Mark Twain


Think about how stupid the average person is; now realise half of them are dumber than that.
- George Carlin
 
This is all could find:http://abcnews.go.com/GMA/AmericanFamily/story?id=1653069&page=1

Good points about statistics.

P.S. a bit more (see item 5)
http://marriage.rutgers.edu/Publications/pubtoptenmyths.htm

5 Leonore J. Weitzman, “The Economics of Divorce: Social and Economic Consequences of Property, Alimony, and Child Support Awards” UCLA Law Review 28 (August, 1981): 1251; Richard R. Peterson, “A Re-Evaluation of the Economic Consequences of Divorce” American Sociological Review 61 (June, 1996): 528-536; Pamela J. Smock, “The Economic Costs of Marital Disruption for Young Women over the Past Two Decades” Demography 30 (August, 1993): 353-371 [back to text]

I wonder if this is available online somewere, at least an abstract...
 
AmadeusM said:
“The Economics of Divorce: Social and Economic Consequences of Property, Alimony, and Child Support Awards”

“A Re-Evaluation of the Economic Consequences of Divorce”

“The Economic Costs of Marital Disruption for Young Women over the Past Two Decades”
Click to expand...
I would gurss there is good information in there. How much of it relates to people in lower, middle, upwardly mobile, and upper income levels? How much relates to people with and without children? I would also guess that the percentages differ significantly for each category.

As a newspaper scare story, it is almost meaningless, if not actually misleading. As serious studies, it means we may be able to learn how to avoid the ptifalls.
 
My graduate work was in survey sampling, I have worked as a statistician for the Mexican government (about two years, until I decided that it was not what I wanted in life).

You can do statistics right and you can do statistics wrong, even if you use proper math and sampling methods.

The easiest way to "lie" with statistics is by using loosely defined terms and playing with their definition, like "standard of living", it seems to relate to economics, health, opportunities, etc., but how is it defined in a mathematical way so that you can use statistics ?, can you change the definition so that you get the results that you want ?.

I remember an exercise where we were told to use census figures to get a predefined number of people "living in misery". We just played around with the definition of "living in misery". Lets start with "people with no fixed income", there's too many, lets add that hey haven't had a job for a year, not enough ?, lets make it two years, lets say that their "dwelling place" (home or whatever) does not have running water or electricity, are we getting close ?, lets add....

If you are a serious researcher make sure that you make and explain your definitions clearly and intelligently, use proper sampling methods and correct math and indicate the calculated margin of error.

If you are not a serious researcher, well... you can say anything you want.

Luis
 
You must log in or register to reply here.
Back
Top