Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Data Manipulation
From
Nick Cox <[email protected]>
To
[email protected]
Subject
Re: st: Data Manipulation
Date
Thu, 7 Apr 2011 10:19:48 +0100
Same answer from me. My ethics preclude me from telling you how to do
something that strikes me as a very bad idea, a dataset for each of
500 persons, but clearly that's no constraint on anybody else.
Otherwise look for -reshape long-. Getting a single date variable
gen date = mdy(act_mm, act_dd, act_yy)
will help lots of Stata things. Your observations are then jointly
identified by -id date-.
On Thu, Apr 7, 2011 at 9:57 AM, Rajaram Subramanian Potty
<[email protected]> wrote:
> This is real data set from a researh study and include arond 500
> persons. In the example, I provided the data of four persons only.
>
> Thanks and regards.
>
> RAM
>
> On Thu, Apr 7, 2011 at 1:59 PM, Nick Cox <[email protected]> wrote:
>> It is not clear to me what you mean by "a data", which I don't think
>> is widely accepted as a standard form of words. I can't imagine it
>> being printed in the Stata Journal, for example.
>>
>> If you want to create different datasets from this master dataset, my
>> guess is that would make most analyses much more difficult, especially
>> one dataset for each person.
>>
>> Some analyses would be easier with a long data structure in which each
>> sex act was a separate observation. That's a -reshape long-.
>>
>> (I set aside the question of whether this is real data or the research
>> ethical!)
>>
>> Nick
>>
>> On Thu, Apr 7, 2011 at 6:55 AM, Rajaram Subramanian Potty
>> <[email protected]> wrote:
>>>
>>> I have one data, each selected persons were followed each day and
>>> enquired about their sex act in each day. For each sex act details
>>> were collected. For example, if the respondent reported 3 acts were
>>> reported, information on all the three sex acts were collected. Now
>>> the data available is for each date. From this data, I wanted to
>>> create a data for each person and also another data for each sex act.
>>> these data set should also include all the variables as well.
>>> Following is an example of the data with few variables for the first 3
>>> sex acts.
>>>
>>> The variable ending with _01, _02, and _03 indicates the 3 sex act in
>>> the same day.
>>>
>>> id act_dd act_mm act_yy doi_dd doi_mm doi_yy tot_act rep_01 rep_02 rep_03 q1_01 q1_02 q1_03
>>> 20020001 7 2 2011 8 2 2011 3 X X X 4 4 3
>>> 20020001 8 2 2011 9 2 2011 1 X 4
>>> 20020001 9 2 2011 10 2 2011 2 X X 3 3
>>> 20020001 10 2 2011 11 2 2011 1 X 1
>>> 20020001 11 2 2011 12 2 2011 0
>>> 20020001 12 2 2011 13 2 2011 2 X X 4 1
>>> 20020001 13 2 2011 14 2 2011 1 X 4
>>> 20020001 14 2 2011 16 2 2011 2 X X 4 4
>>> 20020001 15 2 2011 16 2 2011 2 A A 2 2
>>> 20020001 16 2 2011 17 2 2011 1 X 3
>>> 20020002 12 2 2011 13 2 2011 0
>>> 20020002 13 2 2011 14 2 2011 5 A A X 1 1 4
>>> 20020002 14 2 2011 15 2 2011 2 A A 1 1
>>> 20020002 15 2 2011 16 2 2011 2 A A 4 4
>>> 20020002 16 2 2011 17 2 2011 4 X A A 1 4 4
>>> 20020002 17 2 2011 18 2 2011 2 X X 3 3
>>> 20020003 8 2 2011 9 2 2011 0
>>> 20020003 9 2 2011 10 2 2011 0
>>> 20020003 10 2 2011 11 2 2011 4 A A X 4 4 4
>>> 20020003 11 2 2011 12 2 2011 3 X X X 3 3 4
>>> 20020003 12 2 2011 13 2 2011 0
>>> 20020003 13 2 2011 14 2 2011 2 X X 3 3
>>> 20020003 14 2 2011 15 2 2011 3 X X X 4 3 1
>>> 20020003 15 2 2011 16 2 2011 3 A A X 3 3 3
>>> 20020004 1 3 2011 2 3 2011 0
>>> 20020004 11 2 2011 12 2 2011 1 X 4
>>> 20020004 12 2 2011 15 2 2011 0
>>> 20020004 13 2 2011 15 2 2011 2 X X 3 4
>>> 20020004 14 2 2011 15 2 2011 1 X 3
>>> 20020004 15 2 2011 18 2 2011 0
*
* 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/