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 <[email protected]> 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, [email protected] 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 <[email protected]> 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
[email protected]
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/