Statalist


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

Re: st: missing standard error because of stratum with single sampling unit


From   Michael Begg <mbstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: missing standard error because of stratum with single sampling unit
Date   Wed, 13 Jan 2010 12:10:27 +1100

Steve,

The second stage units are not also stratified. My -svyset- command is:

svyset psu [pw=weight], strata(rural) fpc(npsu) || ssu, fpc(nssu) || hh, fpc(nhh) 

and the singletons appear in ssu. 

Do you have any other suggestions? 

Cheers

Michael 


On 13/01/2010, at 9:58 AM, sjsamuels@gmail.com wrote:

> If I understand you, Michael, the second stage units in your PSUs are
> also stratified.  If those second-stage strata are ordered in any way
> (geographically, for example), I suggest that you assign the singleton
> to a neighboring stratum.
> 
> -Steve
> 
> On Tue, Jan 12, 2010 at 4:51 PM, Michael Begg <mbstata@gmail.com> wrote:
>> Hi Steve,
>> 
>> I have heard back from Stata Technical and it is not a bug.
>> 
>> I made a mistake below, my survey is actually a 3 stage design and the singletons were in the second stage. My apologies for this.
>> 
>> So I guess I now have three options
>> 
>> 1) Drop the observations that are causing the problem - in this case 426 out of a sample of 5,964.
>> 2) Add one of the three singleunit() option to my -svyset- command.
>> 3) Reassign the sampling unit to another appropriately chosen stratum or in my case, reassign the ssu to a different psu.
>> 
>> For 2), does anyone have any recommendations on which option to choose?
>> For 3) how would I go about reallocating the 18 ssu to a new psu?
>> 
>> Cheers
>> 
>> Michael
>> 
>> 
>> On 08/01/2010, at 8:22 AM, sjsamuels@gmail.com wrote:
>> 
>>> It looks like a  bug, and I suggest that you send it to Technical
>>> Support. Be sure to show them (and us) what you see if you run:
>>> 
>>> svydes hhsize, single gen(bad_strat)
>>> 
>>> -Steve
>>> 
>>> On Thu, Jan 7, 2010 at 2:33 AM, Michael Begg <mbstata@gmail.com> wrote:
>>>> Hi,
>>>> 
>>>> I am working on a data from a 2 stage survey design with 2 stratum. When I calculate the mean of hhsize, the standard error is missing as shown below.
>>>> 
>>>> 
>>>> svy: mean hhsize
>>>> (running mean on estimation sample)
>>>> 
>>>> Survey: Mean estimation
>>>> 
>>>> Number of strata =       2        Number of obs    =      5964
>>>> Number of PSUs   =     194        Population size  =  16041309
>>>>                                  Design df        =       192
>>>> 
>>>> --------------------------------------------------------------
>>>>             |             Linearized
>>>>             |       Mean   Std. Err.     [95% Conf. Interval]
>>>> -------------+------------------------------------------------
>>>>      hhsize |   4.656919          .             .           .
>>>> --------------------------------------------------------------
>>>> Note: missing standard error because of stratum with single
>>>>      sampling unit.
>>>> 
>>>> 
>>> 
>>> *
>>> *   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/
>> 
>> 
>> *
>> *   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/
>> 
> 
> 
> 
> -- 
> Steven Samuels
> sjsamuels@gmail.com
> 18 Cantine's Island
> Saugerties NY 12477
> USA
> 845-246-0774
> 
> *
> *   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/


*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index