Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: differences in -svylogitgof- results


From   Imogen Jones <[email protected]>
To   "[email protected]" <[email protected]>
Subject   st: differences in -svylogitgof- results
Date   Mon, 12 Aug 2013 05:07:26 +0000

Hi All

 

I have noticed in the emails lately that some people have been having trouble with -svylogitgof-

 

I have been having a problem that I'm hoping somebody has already figured out an answer to...

 

I am using logistic regression with employment status as my dependent variable (dichotomous) and "multimorbidity" (presence of two or more chronic health conditions) as my independent variable. 

(Multimorbidity is three levels - 0/1=No multimorbidity, 2=2 chronic health conditions, 3=3 or more chronic health conditions).

 

At first, I had my employment status variable coded as 0=employed 1=not employed.  This gave me this output:

 

(Notice the p value of –svylogitgof-)

 

. svy: logistic empstat0 multimorbidity

(running logistic on estimation sample)

 

Survey: Logistic regression

 

Number of strata   =         1                  Number of obs      =      8841

Number of PSUs     =      8841                  Population size    =  16015345

                                                Design df          =      8840

                                                F(   1,   8840)    =    189.94

                                                Prob > F           =    0.0000

 

--------------------------------------------------------------------------------

               |             Linearized

      empstat0 | Odds Ratio   Std. Err.      t    P>|t|     [95% Conf. Interval]

---------------+----------------------------------------------------------------

multimorbidity |   3.491233   .3167115    13.78   0.000     2.922472    4.170683

         _cons |   .4568083   .0154654   -23.14   0.000     .4274766    .4881526

--------------------------------------------------------------------------------

 

. svylogitgof

   Number of observations =                            8841

F-adjusted test statistic = F(1,8840) =               0.172

                 Prob > F =                           0.679

 

 

 

Then when I reversed the coding for my employment status variable to 1=employed 0=not employed, the –svylogitgof- result changed:

 

. svy: logistic empstat multimorbidity

(running logistic on estimation sample)

 

Survey: Logistic regression

 

Number of strata   =         1                  Number of obs      =      8841

Number of PSUs     =      8841                  Population size    =  16015345

                                                Design df          =      8840

                                                F(   1,   8840)    =    189.94

                                                Prob > F           =    0.0000

 

--------------------------------------------------------------------------------

               |             Linearized

       empstat | Odds Ratio   Std. Err.      t    P>|t|     [95% Conf. Interval]

---------------+----------------------------------------------------------------

multimorbidity |   .2864318    .025984   -13.78   0.000     .2397689     .342176

         _cons |   2.189102   .0741126    23.14   0.000      2.04854    2.339309

--------------------------------------------------------------------------------

 

. svylogitgof

   Number of observations =                            8841

F-adjusted test statistic = F(1,8840) =               0.000

                 Prob > F =                           1.000

 

Can anybody tell me why this happens, and more specifically what it means in this instance to get a p=1.0?  Does it simply mean the model is a bad fit, or is there something else going on?

 

Also, when I changed the syntax to show the levels of multimorbidity, this happened:

. svy: logistic empstat0 i.multimorbidity

(running logistic on estimation sample)

 

Survey: Logistic regression

 

Number of strata   =         1                  Number of obs      =      8841

Number of PSUs     =      8841                  Population size    =  16015345

                                                Design df          =      8840

                                                F(   2,   8839)    =    110.26

                                                Prob > F           =    0.0000

 

----------------------------------------------------------------------------------------------

                             |             Linearized

                    empstat0 | Odds Ratio   Std. Err.      t    P>|t|     [95% Conf. Interval]

-----------------------------+----------------------------------------------------------------

              multimorbidity |

        2 Chronic Illnesses  |   4.026233   .4977376    11.27   0.000     3.159772     5.13029

3 or more chronic illnesses  |   8.237624   1.707088    10.18   0.000     5.487604    12.36577

                             |

                       _cons |   .4539598   .0154738   -23.17   0.000     .4246188    .4853283

----------------------------------------------------------------------------------------------

 

. svylogitgof

   Number of observations =                            8841

F-adjusted test statistic = F(1,8840) =               0.000

                Prob > F =                           1.000

 

Does –svylogitgof- not allow for polychotomous IV’s?

 

Thanks in advance!

 

Imogen Jones



© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index