Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: weighted table for multiple response frequency


From   "Ben Jann" <ben.jann@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: weighted table for multiple response frequency
Date   Tue, 14 Nov 2006 22:41:43 +0100

This is because the sum of weights is normalized to the number of
nonmissing cases (as is standard in Stata for aweights) and -mrtab-
considers observation with zero responses as missing by default.
Specify the -include- option to include those observations in the
calculations. In your case, type

 . mrtab ... [aweight=V040101], poly response(934/941) include

and frequencies should be the same.
ben

On 11/14/06, sjyoo <sjyoo@ic.sunysb.edu> wrote:
Dear all:

Working on multiple response table with a weight I have found the
following.
(As you are aware, in a multiple response item respondent can provide
several answers for a question)

First I draw multiple response table (with the command of 'mrtab') of
the whole sample.

. mrtab V043007a V043007b V043007c V043007d V043007e V043059a V043059b
V043059c V043059d V043059e V043013a V043013b V043013c V043013d V043013e
V043055a V043055b V043055c V043055d V043055e [aweight=V040101], poly
response(162/165 511/512 517/518 521/536 544 805/830 847/848 870
900/1097)

            |             Percent of     Percent
            |  Frequency   responses    of cases
------------+-----------------------------------
  162   162 |   2.102727        0.17        0.39
......
  934   934 |   11.80695        0.96        2.20
  935   935 |   27.00409        2.20        5.03
  937   937 |   4.670634        0.38        0.87
  938   938 |   10.22423        0.83        1.90
  939   939 |   26.27211        2.14        4.89
......
 1070  1070 |   1.075062        0.09        0.20
------------+-----------------------------------
      Total |   1229.532      100.00      228.96

Valid cases:        537
Missing cases:      675


Then, I run the same command to see only a subset of the categories.

. mrtab V043007a V043007b V043007c V043007d V043007e V043059a V043059b
V043059c V043059d V043059e V043013a V043013b V043013c V043013d V043013e
V043055a V043055b V043055c V043055d V043055e [aweight=V040101], poly
response(934/941)

            |             Percent of     Percent
            |  Frequency   responses    of cases
------------+-----------------------------------
 934    934 |   10.61642       14.76       17.12
 935    935 |   24.28117       33.76       39.16
 937    937 |   4.199678        5.84        6.77
 938    938 |   9.193286       12.78       14.83
 939    939 |     23.623       32.85       38.10
------------+-----------------------------------
      Total |   71.91355      100.00      115.99

Valid cases:         62
Missing cases:     1150

Why are the numbers of frequency different? Can anybody tell me the
reason?
Thank you so much.

Sung-jin Yoo
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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   |   What's new   |   Site index