Dear All
I need some advice on certain statistical concepts and hoping someone here can provide some direct answers.
I've run a logistic regression model which includes interactions between two categorical variables. I've used Stata's margins command to make sense of the interactions - and the results are both intriguing and interesting (and also according to our hypothesis):
Above, the outcome is 'seekhelp': 0=those who do not seek medical help and 1=those who do My main predictors are 'sexethnic' which indicates an individuals sexual and ethnic identities with four categories and 'comorbid' which is binary that, 0=no comorbidity and 1=having comorbidity.
I used testparm to check for joint effects - the p-value was <0.001
If I understand correctly, testparm is a joint test to assess whether all coefficients associated with the interaction of factor variables 'sexethnic' and 'comorbid' are equal to 0. Since the p<0.05, this tells us that the model with interactions is better than the model without? And hence I should retain the interactions?
Further, what exactly is a joint test?
Further, while I'm not too concerned about statistical significance of the interactions in the model above, the predicted probabilities indicates that the model with interactions makes more sense (theoretically) than the model without interactions.
However, reviewers are ''insisting'' that I need to discuss model fit - that the model with interactions is significantly better than the one without. Generally I would use a likelihood ratio test, but this does not work with survey data.
Thus, is testparm sufficient to say that the model with interactions is better (along with the fact that predicted probabilities from this model also makes sense)?
The predicted probabilities below in case of interest:
. margins (sexethnic#comorbid)
Thanks!
/Amal
I need some advice on certain statistical concepts and hoping someone here can provide some direct answers.
I've run a logistic regression model which includes interactions between two categorical variables. I've used Stata's margins command to make sense of the interactions - and the results are both intriguing and interesting (and also according to our hypothesis):
Code:
svy: logistic seekhelp i.sexethnic##i.comorbid sex i.incomeq margins (sexethnic#comorbid)
Code:
Logistic regression Number of obs = 9,030 LR chi2(12) = 852.35 Prob > chi2 = 0.0000 Log likelihood = -4414.4308 Pseudo R2 = 0.0880 ------------------------------------------------------------------------------------ seekhelp | Odds ratio Std. err. z P>|z| [95% conf. interval] -------------------+---------------------------------------------------------------- sexethnic | White-SM | 3.640788 .2425844 19.39 0.000 3.195068 4.148687 EM-heterosexual | .5976129 .0555956 -5.53 0.000 .4980044 .7171446 EM-SM | 2.752875 .407243 6.85 0.000 2.05999 3.678814 | 1.comorbid | 3.145764 .3903478 9.24 0.000 2.466626 4.01189 | sexethnic#comorbid | White-SM#1 | .6745556 .1313252 -2.02 0.043 .4605771 .9879459 EM-heterosexual#1 | .8079153 .2667794 -0.65 0.518 .4229538 1.543259 EM-SM#1 | .5862236 .253474 -1.24 0.217 .2511984 1.368074 | sex | 1.469779 .0798558 7.09 0.000 1.32131 1.634931 | incomeq3 | 2 | 1.036387 .0871094 0.43 0.671 .8789776 1.221987 3 | .8280044 .0718018 -2.18 0.030 .6985851 .9813998 4 | .8287136 .0724647 -2.15 0.032 .6981895 .9836387 5 | .8039571 .0693916 -2.53 0.011 .6788339 .952143 | _cons | .1301035 .0139224 -19.06 0.000 .1054877 .1604635 ------------------------------------------------------------------------------------
If I understand correctly, testparm is a joint test to assess whether all coefficients associated with the interaction of factor variables 'sexethnic' and 'comorbid' are equal to 0. Since the p<0.05, this tells us that the model with interactions is better than the model without? And hence I should retain the interactions?
Further, what exactly is a joint test?
Further, while I'm not too concerned about statistical significance of the interactions in the model above, the predicted probabilities indicates that the model with interactions makes more sense (theoretically) than the model without interactions.
However, reviewers are ''insisting'' that I need to discuss model fit - that the model with interactions is significantly better than the one without. Generally I would use a likelihood ratio test, but this does not work with survey data.
Thus, is testparm sufficient to say that the model with interactions is better (along with the fact that predicted probabilities from this model also makes sense)?
The predicted probabilities below in case of interest:
. margins (sexethnic#comorbid)
Code:
Predictive margins Number of obs = 9,030 Model VCE: OIM Expression: Pr(seekhelp), predict() --------------------------------------------------------------------------------------- | Delta-method | Margin std. err. z P>|z| [95% conf. interval] ----------------------+---------------------------------------------------------------- sexethnic#comorbid | White-Heterosexual#0 | .1737261 .0052282 33.23 0.000 .1634791 .1839731 White-Heterosexual#1 | .3954843 .0280088 14.12 0.000 .3405881 .4503805 White-SM#0 | .4304796 .0135683 31.73 0.000 .4038861 .457073 White-SM#1 | .6138357 .0329269 18.64 0.000 .5493 .6783713 EM-heterosexual#0 | .1119266 .0083708 13.37 0.000 .09552 .1283331 EM-heterosexual#1 | .2414332 .0536803 4.50 0.000 .1362217 .3466447 EM-SM#0 | .3644274 .0327756 11.12 0.000 .3001885 .4286663 EM-SM#1 | .512133 .0962549 5.32 0.000 .3234768 .7007892 ---------------------------------------------------------------------------------------
/Amal
Comment