Statalist


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

st: RE: patterns of missing data by interviewers


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: patterns of missing data by interviewers
Date   Sun, 22 Jul 2007 16:53:08 +0100

You should give information on, or examples of, your 
variables and data structure. I can't see any here. 

You should also -update- to Stata 9.2.  

Nick 
n.j.cox@durham.ac.uk 

Govind Bell Acharya
 
> It's my first post and I'll try to make this as detailed as 
> possible, so 
> I apologize if this needs clarification:
> 
> For our research, we use telephone interviewers to conduct a 
> number of 
> surveys.   At the moment, it is a challenge to detect whether 
> the number 
> of items where the interviewer coded the missing data (don't know or 
> refused options) is above or below the overall mean of 
> missing values.  
> I did something like that using the proc sql command in SAS, 
> but it is 
> (as SAS is in general), extremely unwieldy and creates major 
> issues such 
> as (f) below.  In any case, here is what I have in mind
> 
> (a)Name   (b)# surveys complete   (c)# missing   (d)# 
> questions asked   
>   (e) (c)/(d)   (f) [sum of (c)]/[sum of (d)]   (g) [(e)-(f)]
> John Doe   12                                 5               
>    120   
>                               5/120            7/200          
>            
>            
> Mary Sue   8                                    2             
>      80   
>                                 2/80            7/200         
>          
>               
> 
> (a) The name of the interviewer;
> (b) number of completions for each interviewer;
> (c) the raw number of missing numbers totaled for each 
> interviewer (so I 
> guess I would code 0 as nonmissing and 1 as missing and sum them by 
> interviewer);
> (d) The total number of questions asked by the interviewer 
> (so if it's a 
> 10 question survey and they completed 8 surveys, they asked 
> 80 questions);
> (e) (c)/(d) is the percentage of missing by interviewer;
> (f) This is the overall mean of missing data across questions 
> asked and 
> across interviewers.  This was something that I have been 
> doing by hand 
> in SAS-- I will run the program, then count the total 
> missing, then go 
> into the code and put those numbers in, which is of course a 
> weird way 
> of doing this;
> (g) the difference between (e) and (f)
> 
> I am pretty well versed in Stata, but an beginner to advanced 
> beginner 
> level in the programming aspects.  I use Stata 9. 

*
*   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