'safer with their dads'

[JohnT]

Che might be on to something there, though I'm not capable of doing the regression analysis needed to determine the correlations:

http://www.census.gov/population/socdemo/hh-fam/cps2002/tabAVG1.xls

"AVERAGE NUMBER OF PEOPLE UNDER 18 IN HOUSEHOLDS:

Married: .91
Male Householder: .85
Female Householder: 1.21"

[Ben Kenobi]

Che:

Yeah, we don't know how many in each, but it seems reasonable that given the size of the sample that the ratio approach unity.

[Ben Kenobi]

Okay, since the only difference in these numbers has to come from families with either a single mom or a single dad, excluding the 'other' category

Male Householder: .85
Female Householder: 1.21

Then the size ratio works out to be 1.42 in favour of the mother.

So, if we have 2x greater chance (or 100%) for the mother's children being abused, with a sample size 1.42x as large:

1.72x as likely or 72% increased chance of being abused by giving the child to the mother, rather than the father.

I'm not sure if that's statistically significant or not.

[JohnT]

It is.

[chequita guevara]

I'm not sure I follow that math. I'm not very good at math any more.