------------------------------------------------------------------------------
log: C:\AAA Miker Files\newer web pages\soc_388_notes\soc_388_2007\Cla
> ss_13_log.log
log type: text
opened on: 8 Nov 2007, 11:32:49
. use "C:\AAA Miker Files\newer web pages\soc_388_notes\soc_388_2007\ed_interm
> ar.dta", clear
. table hed wed, contents(sum count) row col
*See my Comprehensive Excel file for explanations of this....
------------------------------------------------------------
husband's | wife's education
education | <HS HS Some Col BA+ Total
----------+-------------------------------------------------
<HS | 32016 33374 8407 988 74785
HS | 28370 137876 43783 8446 218475
Some Col | 7051 48766 61633 18195 135645
BA+ | 984 13794 28635 51224 94637
|
Total | 68421 233810 142458 78853 523542
------------------------------------------------------------
. desmat: poisson count hed wed
------------------------------------------------------------------------------
Poisson regression
------------------------------------------------------------------------------
Dependent variable count
Optimization: ml
Number of observations: 16
Initial log likelihood: -221501.223
Log likelihood: -113882.425
LR chi square: 215237.595
Model degrees of freedom: 6
Pseudo R-squared: 0.486
Prob: 0.000
------------------------------------------------------------------------------
nr Effect Coeff s.e.
------------------------------------------------------------------------------
count
hed
1 HS 1.072** 0.004
2 Some Col 0.595** 0.005
3 BA+ 0.235** 0.005
wed
4 HS 1.229** 0.004
5 Some Col 0.733** 0.005
6 BA+ 0.142** 0.005
7 _cons 9.187** 0.005
------------------------------------------------------------------------------
* p < .05
** p < .01
. poisgof
Goodness-of-fit chi2 = 227578.9
Prob > chi2(9) = 0.0000
. *Independence model
. set linesize 79
. predict P_indep_again
(option n assumed; predicted number of events)
. table hed wed, contents(sum P_indep_again)
--------------------------------------------------
husband's | wife's education
education | <HS HS Some Col BA+
----------+---------------------------------------
<HS | 9773.551 33398.43 20349.32 11263.7
HS | 28552.2 97569.33 59447.98 32905.5
Some Col | 17727.26 60578.06 36909.58 20430.1
BA+ | 12367.98 42264.19 25751.13 14253.7
--------------------------------------------------
. display (9773.5*97569)/(33398*28552)
1.0000115
. display ln((9773.5*97569)/(33398*28552))
.00001146
. *Now, let's look at the simlplest one term interaction we used before, which is one term for endogamy
. table hed wed, contents(mean ed_endogamy_simple)
--------------------------------------------------
husband's | wife's education
education | <HS HS Some Col BA+
----------+---------------------------------------
<HS | 1 0 0 0
HS | 0 1 0 0
Some Col | 0 0 1 0
BA+ | 0 0 0 1
--------------------------------------------------
. desmat: poisson count hed wed ed_endogamy_simple
-------------------------------------------------------------------------------
Poisson regression
-------------------------------------------------------------------------------
Dependent variable count
Optimization: ml
Number of observations: 16
Initial log likelihood: -221501.223
Log likelihood: -41944.565
LR chi square: 359113.316
Model degrees of freedom: 7
Pseudo R-squared: 0.811
Prob: 0.000
-------------------------------------------------------------------------------
nr Effect Coeff s.e.
-------------------------------------------------------------------------------
count
hed
1 HS 0.740** 0.005
2 Some Col 0.414** 0.005
3 BA+ 0.216** 0.005
wed
4 HS 0.979** 0.005
5 Some Col 0.608** 0.005
6 BA+ 0.081** 0.005
ed_endogamy_simple
7 1 1.115** 0.003
8 _cons 9.067** 0.005
-------------------------------------------------------------------------------
* p < .05
** p < .01
. poisgof
Goodness-of-fit chi2 = 83703.13
Prob > chi2(8) = 0.0000
. *We get a strong measure of endogamy, but it doesn't fit so well...
. * Local table odds ratios from the predicted vals of this model would be exp(1.115) squared along the diagonal, and one elsewhere.
. predict P_diag
(option n assumed; predicted number of events)
. table hed wed, contents(sum P_diag)
--------------------------------------------------
husband's | wife's education
education | <HS HS Some Col BA+
----------+---------------------------------------
<HS | 26426.32 23047.51 15915.36 9395.808
HS | 18145.71 147304.7 33341.21 19683.35
Some Col | 13104.12 34867.67 73458.66 14214.54
BA+ | 10744.85 28590.09 19742.76 35559.3
--------------------------------------------------
. display (exp(1.115))^2
9.2998661
. display 26426.3*147304.7/(23047.5*18145.7)
9.3079794
. *Close enough for government work
. *off the diagonal, our previous model makes no assumptions, so we should have
> local table odds ratio of 1
. display 15915.4*19683.4/(9395.81*33341.2)
1.0000051
. *And there it is.
. *So, the coefficients tell you about odds ratios in the local tables of the p
> redicted values.
. *now let's look at the scores.
. gen score=hed*wed
. table hed wed, contents(mean score)
--------------------------------------------------
husband's | wife's education
education | <HS HS Some Col BA+
----------+---------------------------------------
<HS | 1 2 3 4
HS | 2 4 6 8
Some Col | 3 6 9 12
BA+ | 4 8 12 16
--------------------------------------------------
. desmat: poisson count hed wed @score
-------------------------------------------------------------------------------
Poisson regression
-------------------------------------------------------------------------------
Dependent variable count
Optimization: ml
Number of observations: 16
Initial log likelihood: -221501.223
Log likelihood: -6373.659
LR chi square: 430255.129
Model degrees of freedom: 7
Pseudo R-squared: 0.971
Prob: 0.000
-------------------------------------------------------------------------------
nr Effect Coeff s.e.
-------------------------------------------------------------------------------
count
hed
1 HS -0.836** 0.006
2 Some Col -3.731** 0.012
3 BA+ -7.128** 0.021
wed
4 HS -0.671** 0.006
5 Some Col -3.656** 0.012
6 BA+ -7.418** 0.022
7 score 1.000** 0.003
8 _cons 9.270** 0.005
-------------------------------------------------------------------------------
* p < .05
** p < .01
. *We get one value for score, which turns out by chance to be 1.000.
. poisgof
Goodness-of-fit chi2 = 12561.32
Prob > chi2(8) = 0.0000
. *compared to the one term for endogamy, this one term for linear by linear association fits fairly well.
. predict P_scores
(option n assumed; predicted number of events)
. table hed wed, contents (sum P_scores)
--------------------------------------------------
husband's | wife's education
education | <HS HS Some Col BA+
----------+---------------------------------------
<HS | 28854.76 40075.84 5506.446 347.9548
HS | 33991.05 128324.8 47927 8232.138
Some Col | 5110.256 52440.83 53237.8 24856.12
BA+ | 464.9252 12968.53 35786.75 45416.79
--------------------------------------------------
. display exp(1)
2.7182818
. display 5506.45*8232.14/(347.955*47927)
2.7181982
. *We get constant local table odds ratios.
. *The next obvious question, is whether 1,2,3,4 is the appropriate relative spacing between the categories.
. save "C:\AAA Miker Files\newer web pages\soc_388_notes\soc_388_2007\ed_interm
> ar.dta", replace
file C:\AAA Miker Files\newer web pages\soc_388_notes\soc_388_2007\ed_intermar.
> dta saved
. save "C:\AAA Miker Files\newer web pages\soc_388_notes\soc_388_2007\ed_interm
> ar.dta", replace
file C:\AAA Miker Files\newer web pages\soc_388_notes\soc_388_2007\ed_intermar.
> dta saved
. exit, clear