carry through other fields in a record

[becki]

to continue on with my summarize question, if i summarize -> group by "purchase order" and last "revision", how do I pull through the rest of the fields, like date and amount spent, of the selected records?  the only way I can see, so far, is to put a select on the "command" (what do you call these things, when it's a gui, not a language?) before the summarize, pull out the fields I want to carry through then join to the output of the summarize but is there a way to carry "all" or "rest" of fields through, from a summarize statement?

thanks

becki kain

[gc]

If I understand your question correctly, I think the answer is Yes you would join back your summarized results to the more granular source records on a multi-field key (e.g. purchase order + last revision) that correctly associates the summarized quantities with the associated source records.

[SeanAdams · Accepted solution]

Hi @becki,

There's no capability in the Summarize tool to carry non-summarized rows through - the reason being that when you summarize it reduces the row count and the other rows may be at a different granularity.   So you have to re-join this data after the summarize back to the original data if you want to re-constitute, or else include these fields in the summarize.

Example

Type	Colour	AgeGroup	Weight
Dog	Red	Adult	10
Dog	Grey	Juvenile	5
Cat	Orange	Adult	8
Cat	mottled	Juvenile	4

If I summarize this data by type and weight, I could get 2 rows with the average weight for Dogs & Cats. (which is what I would want)

Type	AvgWeight
Dog	7.5
Cat	6

However, if I tried to pass through colour then you have 4 rows again (because there are 4 distinct values in Colour)

Now - you do have a way to bring Colour, but as a summarised list as follows

Type	AvgWeight	Colours
Dog	7.5	Red, Grey
Cat	6	Orange, Mottled

You can do this using one of the string Concatenate operations in summarize..

Let me know if this helps to answer your question?   If not, perhaps we can work together with some sample data to get you to a result that works.

thanks @becki, have a good weekend

Sean