Assuming you have some id variable, merge on that; there should be
exactly 200 with _merge=3 so just keep those (maybe fastest if you make
dataset2a with just the id variable(s)) as dataset1a (don't forget to
drop _merge)
Rich
On 1/29/13 8:23 AM, Lars Folkestad wrote:
> That was my idea too.
> But how?
>
> lars
>
> Den 29/01/13 14.18 skrev "Richard Goldstein" <richgold@ix.netcom.com>:
>
>> why not just make a new data set (say "dataset1a") consisting only of
>> the 200?
>>
>> Rich
>>
>> On 1/29/13 8:14 AM, Lars Folkestad wrote:
>>> Dear List
>>>
>>> I have to data sets
>>>
>>> dataset1: Consists of 40 variables and 800 unique observations
>>> (persons).
>>> dataset2: consists of the same 40 variables and 200 randomly selected
>>> from
>>> dataset 1 entered to be able to check for concistancy in entry.
>>>
>>> All data are based on data from a questionnaire.
>>>
>>> I want to know if there are many mis matches.
>>>
>>> I am going to use the following command:
>>>
>>> .use dataset1, clear
>>> .cf _all using dataset2, verbose
>>>
>>>
>>>
>>> But how do i get stata to only compare the 200 sets of questionnaire
>>> data
>>> that was double entered?
>>>
>>> Thank you
>>>
>>> Lars
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
