[Babase] Re: agonism/grooming 'investigation'

Karl O. Pinc kop at meme.com
Sun Mar 25 16:24:31 EDT 2007


On 03/25/2007 02:06:04 PM, Susan Alberts wrote:
> Sorry to be slow getting back to you but i wasn't completely sure I  
> understood your message this morning and didn't have time to think it  
> through.
> 
> I have copied this reply to the babase mailing, so that the  
> discussion will be in the archive.
> 
> Responses below.
> 
> 
>>  I wanted to know if you think I should go in and look
>> at the years where it looks like they WERE pushed to the first to  
>> see how much
>> so each month, etc...
> 
> 
> No I don't think this is very important. That is, in years where  
> there is clearly not an even distribution of events across days  
> (where it was usually pushed to 1st of month) we don't need to know  
> much more about those years -- we can't recover the "real" dates  
> anyway (not very easily). So the years where it is clear that most  
> were pushed to the first are not the issue. I think the issue is, do  
> we have a clear sense of which year the date of the entry actually  
> starts meaning the real date. In other words, if we leave the dates  
> as they are for a given year, can we be sure that they really  
> represent the real dates, or were some pushed to the first of the  
> month. To think about this, I think that you should pick the first  
> reasonably "even" year for each of grooming and agonism and then plot  
> the events-by-day on a bar graph so we can see the distribution. You  
> can ignore the years where things were mostly pushed to the first.

select act
     , extract(year from date) as year_part
     , extract(month from date) as month_part
     , sum(case when extract(day from date) = 1 then 1 else 0 end) as  
_1st
     , sum(case when extract(day from date) = 2 then 1 else 0 end) as  
_2nd
     , sum(case when extract(day from date) = 3 then 1 else 0 end) as  
_3rd
     , sum(case when extract(day from date) = 4 then 1 else 0 end) as  
_4th
     , sum(case when extract(day from date) = 5 then 1 else 0 end) as  
_5th
     , sum(case when extract(day from date) = 6 then 1 else 0 end) as  
_6th
     , sum(case when extract(day from date) = 7 then 1 else 0 end) as  
_7th
     , sum(case when extract(day from date) = 8 then 1 else 0 end) as  
_8th
     , sum(case when extract(day from date) = 9 then 1 else 0 end) as  
_9th
     , sum(case when extract(day from date) = 10 then 1 else 0 end) as  
_10th
     , sum(case when extract(day from date) = 11 then 1 else 0 end) as  
_11th
     , sum(case when extract(day from date) = 12 then 1 else 0 end) as  
_12th
     , sum(case when extract(day from date) = 13 then 1 else 0 end) as  
_13th
     , sum(case when extract(day from date) = 14 then 1 else 0 end) as  
_14th
     , sum(case when extract(day from date) = 15 then 1 else 0 end) as  
_15th
     , sum(case when extract(day from date) = 16 then 1 else 0 end) as  
_16th
     , sum(case when extract(day from date) = 17 then 1 else 0 end) as  
_17th
     , sum(case when extract(day from date) = 18 then 1 else 0 end) as  
_18th
     , sum(case when extract(day from date) = 19 then 1 else 0 end) as  
_19th
     , sum(case when extract(day from date) = 20 then 1 else 0 end) as  
_20th
     , sum(case when extract(day from date) = 21 then 1 else 0 end) as  
_21st
     , sum(case when extract(day from date) = 22 then 1 else 0 end) as  
_22nd
     , sum(case when extract(day from date) = 23 then 1 else 0 end) as  
_23rd
     , sum(case when extract(day from date) = 24 then 1 else 0 end) as  
_24th
     , sum(case when extract(day from date) = 25 then 1 else 0 end) as  
_25th
     , sum(case when extract(day from date) = 26 then 1 else 0 end) as  
_26th
     , sum(case when extract(day from date) = 27 then 1 else 0 end) as  
_27th
     , sum(case when extract(day from date) = 28 then 1 else 0 end) as  
_28th
     , sum(case when extract(day from date) = 29 then 1 else 0 end) as  
_29th
     , sum(case when extract(day from date) = 30 then 1 else 0 end) as  
_30th
     , sum(case when extract(day from date) = 31 then 1 else 0 end) as  
_31st
            from interact_data
           group by act, year_part, month_part
           order by act, year_part, month_part;


Karl <kop at meme.com>
Free Software:  "You don't pay back, you pay forward."
                  -- Robert A. Heinlein



More information about the Babase mailing list