Econlib Resources
|
From Will Wilkinson, in a 100+ comment discussion at Overcoming Bias:
I would guess that number of children is negatively correlated to number of sexual partners.According to the General Social Survey, Will's guess is... correct! The correlation between number of children and PARTNRS5, its best measure of total number of sexual partners, is -.26. This result holds up in a multiple regression that controls for age and the year in which the question was asked:  Nice guesswork, Will. * Note: Here's how the GSS codes responses for PARTNRS5. To make the variable monotonic, I only counted responses from 0-8.
CATEGORIES:
Family Economics
|
||||||||
|
Blogging software: Powered by Movable Type 4.2.1.
Pictures courtesy of the authors. All opinions expressed on EconLog reflect those of the author or individual commenters, and do not necessarily represent the views or positions of the Library of Economics and Liberty (Econlib) website or its owner, Liberty Fund, Inc. 
The cuneiform inscription in the Liberty Fund logo is the
earliest-known written appearance of the word
"freedom" (amagi), or "liberty." It
is taken from a clay document written about 2300 B.C. in the Sumerian city-state of Lagash.
|
||||||||
Isn't there a logical problem here?
If we assume that those who are married are less likely than those who are not to have had multiple partners over the last five years, aren't you really just showing that if one is married one is more likely to have more children than if one is single?
Maybe age serves as a proxy here, but not necessarily. Is there a way to control for marital status?
(You already pointed out the fact that this particular variable shows only a five year window, so doesn't really measure total sexual partners. I don't think controlling for age fixes that problem.)
Why are you impressed? That seems pretty obvious to me and probably many alphas of the world.
Is the database restricted to heterosexuals? Is the database restricted to men (as I could easily imagine women with more sexual partners having more children)?
Is the database restricted to heterosexuals? Is the database restricted to men (as I could easily imagine women with more sexual partners having more children)?
don't ask bryan to do your work for you, do it yourself:
http://sda.berkeley.edu/cgi-bin/hsda?harcsda+gss08
but last i heard it applied to women too.
I would guess that number of children you don't know about is positively correlated with number of partners.
Audacious Epigone and Inductivist wrote about this a while back.
Epigone also looked at women. And Inductivist also looked at trends over time.
I would guess that number of abortions is positively correlated to number of sexual partners.
Sex with multiple partners = more STDs and more pregnancies.
Pregnancies + no steady parters = more abortions and fewer births.
I did my OB/GYN clinical clerkship in a Brooklyn neighborhood that was becoming a slum. I don't need survey data to know that the above statements are correct.