Re: st: Deleting Duplicates based on criteria

[Katie Farrin <kfarrin@gmail.com>]

Could you change your dummies into a single variable, with the most severe
charge taking the a value of 1 and lesser charges taking higher values?
Then you could sort by these values and run:

sort caseid charge

quietly by caseid:  gen dup = cond(_N==1,0,_n)

drop if dup>1

I'm sure there are more efficient ways to do this but just a suggestion.

Katie

On Thu, Jul 11, 2013 at 10:43 AM, Dirlam, Jonathan C.
<dirlam.1@buckeyemail.osu.edu> wrote:
> Highest charge determined by this order: 1. Homicide, 2. Sex offense, 3. Robbery, 4. Agg Assault, 5. Drug Trafficking, 6. Burglary, 7. Larceny Theft, 8. Motor Vehicle Theft, 9. Drug Sales, 10. Weapon, 11. DUI, 12. Drug Possession, 13. Other
>
> Example of data with 3 of 13 dummies:
> Court case number        id        robberydummy    burglarydummy    homicidedummy
> 000000038CFMA            6                 1                           0                         0
> 000000038CFMA            6                 1                           0                         0
> 000000038CFMA            6                 0                           1                         0
> 000000045CFMA            8                 1                           0                         0
> 000000045CFMA            8                 0                           0                         1
>
> In this example, I want one of the robbery observations for id=6 and the homicide observation for id=8.
> Thanks.
>
> ________________________________________
> From: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu] on behalf of Nick Cox [njcoxstata@gmail.com]
> Sent: Thursday, July 11, 2013 10:23 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: Deleting Duplicates based on criteria
>
> Yes,  but tell us the rules for determining the highest charge and
> give us a realistic example of a block of observations for some court
> case. (Need not be real, just realistic.)
> Nick
> njcoxstata@gmail.com
>
>
> On 11 July 2013 15:18, Dirlam, Jonathan C. <dirlam.1@buckeyemail.osu.edu> wrote:
>> Dear Statalist,
>> I have duplicate observations where the duplicates are the same court case number. I want to eliminate all the observations for a court case except for the observation that has the highest charge (homicide, robbery, etc.) I have 12 dummy variables that capture charges and used the duplicates command to get unique ids for each court case number. Is there a way to write a program that eliminates or keeps duplicates based on criteria you give it (Example, homicidedummy==1) and stops once all but one observation are eliminated?
>> Thanks.
>>
>>
>> *
>> *   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/
>
> *
> *   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/
>
>
>
>
>
> *
> *   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/

*
*   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/