Re: st: creating random groups of observations

[Luca Campanelli <l.campanelli@yahoo.it>]

thank you, John, for your answer. 

I think that it would not work because I don't want each word to be in more than one group at the same time. 

Luca




John Luke Gallup <jlgallup@pdx.edu> wrote: 

Luca,

You can generate a lot more than a 1000, and delete those that don't meet your criteria.

John




On Dec 5, 2012, at 5:00 PM, Luca Campanelli <l.campanelli@yahoo.it> wrote:


> I apologize for my previous email with no subject. 
> Dear Stata users, 
> I have a string variable "word" with 4000 words, and I'd like to create 1000 groups of 4 randomly selected words (the group of words will be trials in an experimental task). I also have info about the number of characters of each word. 
> 
> The dataset: 
>    ID   WORD      CHR
>    1    acorn     5
>    2    address   7
>    3    adult     5
> 
> One way to do it could be: 
>    generate rdm = runiform()
>    sort rdm
>    egen group = seq(), from(1) to(1000) block(1)
> 
> My question is if I can constrain the groups to meet a certain criterion: I'd like that each group of words has a total number of characters between given limits, for example between 18 and 28 total number of characters. I mean that I don't want groups with 4 short words or groups with 4 long words. 
> Do you know if there is any way to do it? 
> 
> Thank you for any help anybody can provide.
> Luca Campanelli
> 
> PS I have Stata IC 12.1 on Windows 7. 


*
*   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/