Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

st: post hoc test categorical survey data


From   <priscilla.martinez@medisin.uio.no>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: post hoc test categorical survey data
Date   Wed, 24 Aug 2011 12:03:06 +0200

Hello,

I'm working with survey data, and I have a nominal variable on alcohol use
that I have tabulated with various categorical demographic variables (some
with 2 categories, some with more). Here is an example:

. svy, subpop(age50p): tab age3cats qftype4 if q1009==2 & country==1, column
obs ci se
(running tabulate on estimation sample)

Number of strata   =        18                  Number of obs      =
2401
Number of PSUs     =       238                  Population size    =
5784867.8
                                                Subpop. no. of obs =
2052
                                                Subpop. size       =
1367537.7
                                                Design df          =
220

----------------------------------------------------------------------------
------------------------
          |                                         qftype4

 age3cats |      lifetime       previous       12 month       low risk
at risk          Total
----------+-----------------------------------------------------------------
------------------------
    50-64 |         .5144           .494          .5793          .5977
.6896          .5399
          |       (.0189)        (.0325)        (.0263)        (.0369)
(.058)        (.0137)
          | [.4772,.5514]  [.4303,.5579]  [.5268,.6301]  [.5234,.6678]
[.5656,.7913]  [.5129,.5667]
          |           519            148            193            123
48           1031
          |
      65+ |         .4856           .506          .4207          .4023
.3104          .4601
          |       (.0189)        (.0325)        (.0263)        (.0369)
(.058)        (.0137)
          | [.4486,.5228]  [.4421,.5697]  [.3699,.4732]  [.3322,.4766]
[.2087,.4344]  [.4333,.4871]
          |           568            160            161            103
29           1021
          |
    Total |             1              1              1              1
1              1
          |

          |

          |          1087            308            354            226
77           2052
----------------------------------------------------------------------------
------------------------
  Key:  column proportions
        (linearized standard errors of column proportions)
        [95% confidence intervals for column proportions]
        number of observations

  Pearson:
    Uncorrected   chi2(4)         =   21.7610
    Design-based  F(3.93, 865.47) =    3.7270     P = 0.0054

I'd like to run a post-hoc test to identify between which groups the
significant difference lies. I've read the FAQ on the STATA website and it
describes how to do so with a oneway tab and continuous variables, and I've
scoured other sites and the stata manual and have been unable to find a
description of how to do this. Is this even possible? I'd appreciate any
assistance you can give me.

Many thanks,
Priscilla


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index