Yes, this is a T/J. I have done it because statistics are regularly misused, and the misuse hurts people who are trying to understand the world.
This debate is irrelevant to waitedwaytoolong. I'm not worried about waited - he seems to understand SI's rule 1 - accept what makes sense, and reject the rest.
I am concerned about MrSpock and, now, anothermr, who speak with utter confidence in their positions, based in part on what appears to be statistical evidence, and their statistical evidence is more than questionable.
***********************************************
anothermr:
You know, those things don't go together, right?
Umm...I see a big difference between bias and randomness. The Literary Digest poll of 1936, which predicted a landslide for Alf Landon, failed because it wasn't random, but it's question was pretty unbiased.
A good random sample of US adults can be asked biased questions. The results will be garbage. For example, 'Which tastes better - this great tasting candy or this yucky vegetable? We'll give you a case of the candy if you choose it.'
The answers of a self-selected group of individuals who visit a site named www.dearpeggy.com can't be used to extrapolate to any other demographic.
1) PV's responders did not find her site and then answer surveys. 2) Every study of infidelity that I've seen except the General Social Survey uses self-selected respondents. 3) No one with any basic knowledge of statistics thinks or argues that PV's data is valid for anything but the responders.
Her conclusions are suggestive, though.
given the frequency of divorce, 33% of all divorces for any given year for which we have data would exceed the 50% mark of any credible annual infidelity rate I have seen published, so it would appear that most marriages do not survive infidelity.
You seem to be saying 33% in a given year is more than a 50% annual rate. IOW, a 33% annual rate equals a 50% annual rate. If you come back to this, please show us all your math. Or did you not say what you mean?
Much more important, though, is the fact the percentage of Ds in which infidelity is cited as a cause says nothing about the percentage of Ms with infidelity end in D.
Here's the word problem:
There are 1,200,000 annual Ds in the US. 400,000 of them cite infidelity as one cause. How many US Ms experience infidelity?
The data you've given data seem insufficient for solving it.
Amato, Paul R., and Stacy J. Rogers. 1997. “A Longitudinal Study of Marital Problems and Subsequent Divorce.” Journal of Marriage and the Family 59(3):612–24.
My budget doesn't allow me to spend $20 on this. I'd have to give up higher priority items to buy the article. But the study shows
'...infidelity has been shown to be related to increased marital distress, conflict, and divorce (Amato & Rogers, 1997.'
Not shocking to any member of SI....
************************************************
MrSpock:
I trust you realize that simply repeating your statistics is not at all the same as citing sources, and for the most part, all you did was repeat yourself.
I searched on ‘katie crown divorce’, ‘kate crown divorce’, and ‘katherine crown divorce’ and found noting relevant – a Canadian actress, a singer, Kate Middleton, Katie Holmes – no articles or books on divorce or infidelity.
A search on ‘Holmes infidelity divorce’ brings up stuff like a Houston family law practice, more Katie Holmes … but not articles or books with statistics. Please provide something more than ‘holmes’; even a first name might help.
I look forward to getting more info on the NYT article(s) - when the report on a study, they usually point the reader to the actual study.
If NYP mean the NY Post, skip it - any relation between what the Post says and reality is coincidental, as far as I can tell.
Thank you for specifying 'Infidelity and Behavioral Couple Therapy: Relationship Outcomes Over 5 Years Following Therapy' as a source. I am surprised, however, at the weight you give it.
First, the article came from a study comparing the effects of 2 types of MC. That is, the authors did one study and then picked out the couples who experienced infidelity. This is bass-ackwards, if the goal is to study infidelity.
Second, the number of couples experiencing infidelity was 19! 19. I have a hard time getting over that. To call any conclusions coming from a sample this small is to stretch the meaning of 'suggestive'. Sure, it's a data point, and the more data points the better, but 19? That's awfully small.
Third, the 80% D rate for 'secret infidelity' means 4 out of 5 couples D'ed. If 19 is small, 5 is simply too small to pay any attention to.
Very gently, MrSpock, I think you quote this 80% D rate only because it supports your gut. I cannot imagine any other reason for going through your Procrustean manipulation of this article.
***************************************************************************
Here's the thing: statistics are meaningless to individual cases. It doesn't matter to anyone how many people D after infidelity as long that person is in the group he wants to be in.
But a number of members believe that the majority of men who are betrayed D, and they push those beliefs on others.
I trust readers who have followed this thread realize that those folks have provided no usable data to back up their beliefs. Unless they do, I hope no one buys into anything but uncertainty about the rate of D among betrayed men (and women).
[This message edited by sisoon at 5:19 PM, August 21st (Sunday)]