Basic Modeling Question

Question

Hi,

I have pretty extensive knowledge of ArcGIS 9.2, GUI-wise, but very little experience with coding. Here's my very simple problem, which I hope an equally simple model can fix:

I have an excel file that contains geocodable addresses. (no problem here.) My street file uses the Zip Code included in the excel file in order to geocode.  The excel file also contains a seperate column that contains numbers amounts ($'s) for each address.

I have a separate dbf file (though I could save it as an excel or visa versa) that includes different number amounts (also $'s.) The dbf file only contains STFID's (census tract) for each entry.

So, I can map each file without a problem, creating two separate layers w/ dot density, color gradient etc. I need to count the # (dollar) properties of each address that lie within a given census tract, yet I only have the zip code's for each address. I can achieve this for each individual census tract manually, by highlighting a single census tract and then selecting by location. But for the area I'm mapping (New York City) there are hundreds of census tracts. I need to create a model that can count the attributes of each zip code-geocoded location (again, the $ amounts associated with them) and aggregate them within census tracts, eventually creating a report that allows me to draw a ration between the #'s I already have for each census tract and the aggregate #'s I hope to draw from this hypothetical model.

I'm reading up on Python, the modelling app and the appropriate toolboxes, but if anyone has had experience with this relatively simple issue, pointing me in the right direction - even if that only means which toolbox to utilize - would be greatly appreciated.

Thanks!

SamuelOakford · Answer

I am not sure I understand what you are exactly trying to do, but a spatial join is a good way to aggregate and summarize point data within polygons.  By default you can run some basic statistics (Sum, min, max, etc) on all numeric fields when you join the points to the polygons.  If you want to select only certain fields to summarize you can spatial join the polygon to the points then run summary statistics on the census track name field that was joined over.  You can do both of these via geo-processing so you can stick them in a model and just have it be one step.

Thanks, I'll try that. Perhaps this clearer:

I have two spreadsheets:
Spreadsheet 1) Column A has addresses, column B has zip codes, column C has numeric values x. These addresses are geocoded by zip and address and obviously, I can do a graduated symbol map etc with the X values.
Spreadsheet 2) Column A has a census tracts, column  B has numeric values Y.

I can plot all aspects of both in ArcGIS.

I'd like to have numeric values X (from spreadsheet 1 and geocoded by zip) joined to the spreadsheet 2, to create a column that aggregates all X value's that lie within a given census tract (which can be done individually using select by location.) If spreadsheet 1 had census tract data for every address, then I could do a simple join. But it doesn't, so I need a way to aggregate the X values that lie in each census tract.

Hope that helps,

thanks again!

RustyRex · Answer

I am not sure I understand what you are exactly trying to do, but a spatial join is a good way to aggregate and summarize point data within polygons.  By default you can run some basic statistics (Sum, min, max, etc) on all numeric fields when you join the points to the polygons.  If you want to select only certain fields to summarize you can spatial join the polygon to the points then run summary statistics on the census track name field that was joined over.  You can do both of these via geo-processing so you can stick them in a model and just have it be one step.