I relayed that claim from Greg Cochran a while back, but inspired by this Half Sigma post I decided to present the actual numbers from the GSS.
Row: CHILDS
Column: WORDSUM
Control: SEX
Summary Statistics for SEX = 1(MALE) | ||||||||
---|---|---|---|---|---|---|---|---|
Eta* = | .06 | Gamma = | -.02 | Rao-Scott-P: F(80,38080) = | 2.10 | (p= 0.00) | ||
R = | -.04 | Tau-b = | -.01 | Rao-Scott-LR: F(80,38080) = | 2.02 | (p= 0.00) | ||
Somers’ d* = | -.01 | Tau-c = | -.01 | Chisq-P(80) = | 213.72 | |||
Chisq-LR(80) = | 206.45 | |||||||
*Row variable treated as the dependent variable. |
Summary Statistics for SEX = 2(FEMALE) | ||||||||
---|---|---|---|---|---|---|---|---|
Eta* = | .12 | Gamma = | -.08 | Rao-Scott-P: F(80,38080) = | 4.34 | (p= 0.00) | ||
R = | -.11 | Tau-b = | -.06 | Rao-Scott-LR: F(80,38080) = | 3.85 | (p= 0.00) | ||
Somers’ d* = | -.06 | Tau-c = | -.06 | Chisq-P(80) = | 485.56 | |||
Chisq-LR(80) = | 430.81 | |||||||
*Row variable treated as the dependent variable. |
Summary Statistics for all valid cases | ||||||||
---|---|---|---|---|---|---|---|---|
Eta* = | .08 | Gamma = | -.05 | Rao-Scott-P: F(80,38080) = | 4.70 | (p= 0.00) | ||
R = | -.08 | Tau-b = | -.04 | Rao-Scott-LR: F(80,38080) = | 4.33 | (p= 0.00) | ||
Somers’ d* = | -.04 | Tau-c = | -.04 | Chisq-P(80) = | 522.38 | |||
Chisq-LR(80) = | 481.45 | |||||||
*Row variable treated as the dependent variable. |
I’ve come across that “Firepower” character in the comments section at Overcoming Bias, there as well he seemed quite sure of himself in the absence of supporting evidence.
At Sister Y’s suggestion I am showing some actual tables. Showing all the columns would take up more space than wordpress easily permits, so I have grouped buckets with WORDSUM(r: “BOTTOM” 0-4; “LOWER-MIDDLE” 5; “MIDDLE” 6; “UPPER-MIDDLE” 7; “TOP” 8-10).
First men:
Statistics for SEX = 1(MALE) | |||||||
---|---|---|---|---|---|---|---|
Cells contain: –Column percent -Weighted N |
WORDSUM | ||||||
1 BOTTOM |
2 LOWER-MIDDLE |
3 MIDDLE |
4 UPPER-MIDDLE |
5 TOP |
ROW TOTAL |
||
CHILDS | 0 | 31.1 865.9 |
32.0 603.8 |
34.6 843.6 |
33.5 601.6 |
32.4 863.0 |
32.6 3,778.0 |
1 | 16.8 468.4 |
14.5 272.7 |
14.8 361.6 |
13.3 239.6 |
13.5 360.3 |
14.7 1,702.7 |
|
2 | 20.6 575.1 |
24.2 457.3 |
23.3 567.1 |
23.9 429.0 |
26.0 692.0 |
23.5 2,720.6 |
|
3 | 13.0 361.8 |
13.5 255.0 |
14.9 363.1 |
15.1 272.2 |
14.9 397.5 |
14.3 1,649.6 |
|
4 | 8.4 235.4 |
7.5 140.7 |
6.6 160.5 |
8.0 144.7 |
7.4 197.1 |
7.6 878.3 |
|
5 | 4.3 118.8 |
4.3 80.2 |
3.0 73.0 |
3.8 67.8 |
2.8 74.4 |
3.6 414.1 |
|
6 | 2.4 67.0 |
2.0 37.8 |
1.2 28.7 |
1.0 18.5 |
1.4 37.6 |
1.6 189.6 |
|
7 | 1.5 41.6 |
.6 12.1 |
.7 17.8 |
.9 15.3 |
1.0 26.4 |
1.0 113.3 |
|
8: EIGHT OR MORE | 1.9 54.0 |
1.4 26.3 |
.9 21.3 |
.5 9.5 |
.6 14.8 |
1.1 125.8 |
|
COL TOTAL | 100.0 2,788.0 |
100.0 1,886.0 |
100.0 2,436.7 |
100.0 1,798.1 |
100.0 2,663.1 |
100.0 11,571.9 |
|
Means | 1.92 | 1.82 | 1.67 | 1.74 | 1.74 | 1.78 | |
Std Devs | 1.93 | 1.79 | 1.67 | 1.67 | 1.65 | 1.75 | |
Unweighted N | 2,659 | 1,784 | 2,308 | 1,664 | 2,635 | 11,050 |
Then women:
Statistics for SEX = 2(FEMALE) | |||||||
---|---|---|---|---|---|---|---|
Cells contain: –Column percent -Weighted N |
WORDSUM | ||||||
1 BOTTOM |
2 LOWER-MIDDLE |
3 MIDDLE |
4 UPPER-MIDDLE |
5 TOP |
ROW TOTAL |
||
CHILDS | 0 | 20.5 587.5 |
22.2 519.3 |
21.9 702.5 |
25.6 599.1 |
27.1 930.7 |
23.6 3,339.0 |
1 | 18.0 515.7 |
16.4 383.6 |
16.9 541.9 |
15.2 354.8 |
15.0 514.0 |
16.3 2,310.0 |
|
2 | 21.0 602.0 |
26.1 609.2 |
28.0 894.9 |
26.7 624.7 |
27.8 954.0 |
26.0 3,684.9 |
|
3 | 15.4 441.9 |
17.0 397.4 |
18.0 575.1 |
16.2 379.6 |
16.2 556.0 |
16.6 2,349.9 |
|
4 | 9.9 283.8 |
9.4 219.1 |
8.2 263.4 |
8.6 201.5 |
8.3 286.4 |
8.8 1,254.3 |
|
5 | 5.8 165.1 |
3.6 84.4 |
3.6 115.4 |
3.7 86.3 |
2.9 100.0 |
3.9 551.3 |
|
6 | 3.1 88.1 |
2.1 49.4 |
1.7 54.2 |
1.9 43.5 |
1.1 38.7 |
1.9 274.0 |
|
7 | 2.3 65.1 |
1.1 25.0 |
.7 23.4 |
1.2 29.0 |
.7 23.3 |
1.2 165.8 |
|
8: EIGHT OR MORE | 4.2 120.0 |
2.1 47.9 |
1.0 31.0 |
.8 18.3 |
.8 28.7 |
1.7 246.0 |
|
COL TOTAL | 100.0 2,869.2 |
100.0 2,335.4 |
100.0 3,201.8 |
100.0 2,336.9 |
100.0 3,431.9 |
100.0 14,175.2 |
|
Means | 2.42 | 2.12 | 2.01 | 1.96 | 1.85 | 2.07 | |
Std Devs | 2.11 | 1.80 | 1.64 | 1.71 | 1.62 | 1.79 | |
Unweighted N | 3,059 | 2,371 | 3,226 | 2,336 | 3,531 | 14,523 |
Now everybody:
Statistics for all valid cases | |||||||
---|---|---|---|---|---|---|---|
Cells contain: –Column percent -Weighted N |
WORDSUM | ||||||
1 BOTTOM |
2 LOWER-MIDDLE |
3 MIDDLE |
4 UPPER-MIDDLE |
5 TOP |
ROW TOTAL |
||
CHILDS | 0 | 25.7 1,453.3 |
26.6 1,123.1 |
27.4 1,546.0 |
29.0 1,200.8 |
29.4 1,793.7 |
27.6 7,117.0 |
1 | 17.4 984.0 |
15.5 656.3 |
16.0 903.5 |
14.4 594.4 |
14.3 874.4 |
15.6 4,012.6 |
|
2 | 20.8 1,177.1 |
25.3 1,066.5 |
25.9 1,462.1 |
25.5 1,053.7 |
27.0 1,646.1 |
24.9 6,405.4 |
|
3 | 14.2 803.7 |
15.5 652.5 |
16.6 938.1 |
15.8 651.8 |
15.6 953.4 |
15.5 3,999.5 |
|
4 | 9.2 519.2 |
8.5 359.8 |
7.5 423.9 |
8.4 346.2 |
7.9 483.5 |
8.3 2,132.6 |
|
5 | 5.0 283.9 |
3.9 164.6 |
3.3 188.4 |
3.7 154.1 |
2.9 174.4 |
3.7 965.4 |
|
6 | 2.7 155.1 |
2.1 87.2 |
1.5 82.9 |
1.5 62.0 |
1.3 76.3 |
1.8 463.5 |
|
7 | 1.9 106.8 |
.9 37.1 |
.7 41.2 |
1.1 44.3 |
.8 49.7 |
1.1 279.1 |
|
8: EIGHT OR MORE | 3.1 174.0 |
1.8 74.3 |
.9 52.3 |
.7 27.8 |
.7 43.5 |
1.4 371.8 |
|
COL TOTAL | 100.0 5,657.2 |
100.0 4,221.4 |
100.0 5,638.6 |
100.0 4,135.0 |
100.0 6,094.9 |
100.0 25,747.1 |
|
Means | 2.18 | 1.99 | 1.86 | 1.87 | 1.80 | 1.94 | |
Std Devs | 2.04 | 1.80 | 1.66 | 1.69 | 1.64 | 1.78 | |
Unweighted N | 5,718 | 4,155 | 5,534 | 4,000 | 6,166 | 25,573 |
June 21, 2012 at 12:26 pm
I wonder how the sex difference in IQ/fecundity changes throughout the distribution. I suspect it’s especially pronounced at the top and bottom (bigger gap men>women at the top, bigger gap women>men at the bottom). No time to check now though – nothing readily googlable.
June 21, 2012 at 2:40 pm
you left out a zero at halfsigma
June 22, 2012 at 9:34 pm
namae nanka, thanks for pointing that out. I have responed in the comments at HS, though perhaps a bit too late.
Sister Y, your wish is my command.
June 26, 2012 at 12:24 pm
Fascinating that it’s the top end of the WORDSUM scale where the differences show up, hardly at all at the bottom.
July 15, 2012 at 8:31 am
Women have more children on average than men? How does that work? Is it just response bias, or is there something here that I’m missing?
July 16, 2012 at 12:11 am
Response bias is a definite probability, but it’s not logically impossible for women to KNOW about more of their kids than men.