I know how to join things, but only when they both have matching fields. I have created my own fields before as well, but only when I have had small datasets (less than 100), so it was practical for me to fill in each field by hand.

Now I have the parcelbase data for Saint Louis City (over 128,000 parcels), and the March 2017 crime .csv data from SLMPD (approx 4,000 crimes in March 2017). Both the parcelbase and SLMPD data have addresses, but they have them in different formats (one has street and the other has street direction and street - ex: Field St vs N Field St), so I can't join them since they won't match up in Validate. The second problem is the addresses are two fields each - the address number (ex, 5400) and the street name (ex, Field St), and I don't know how to join based on matching two fields instead of just one.

My boss thinks this isn't something that can be done with Join at all (and I agree), but neither of us know how to match them up otherwise. I do have X Y coordinates for the SLMPD data (not for the parcelbase, but the parcelbase is projected to _1927 and the SLMPD is just an attribute table from a .csv file).

Is this a Georeferencing thing?

How do you do one of those?

If it is not a Georeferencing issue, does anybody know what it is that I need to do to combine these two things?

We want to visually map crime in Saint Louis City for March 2017.

  • parcel numbers should match, you said its parcel based. join with the parcel field.
    – NULL.Dude
    Apr 24, 2017 at 19:17
  • I did go to Geocoding > Geocode Address > World Geocode Service (Add) > but it wouldn't give me the option of adding the .csv file. Do I have to convert it to something?
    – T. Kerr
    Apr 24, 2017 at 19:18
  • The SLMPD data does not include parcel numbers, so there are no parcel numbers to match to.
    – T. Kerr
    Apr 24, 2017 at 19:19
  • What are you trying to do? Do you need your crime data linked to a parcel, or are you just trying to get it from your CSV into a map?
    – Midavalo
    Apr 24, 2017 at 19:39
  • Welcome to GIS SE! As a new user please take the tour to learn about our focused Q&A format.
    – Midavalo
    Apr 24, 2017 at 19:39

1 Answer 1


Joining based on address does not usually work because addresses from difference sources are almost always different, different abbreviations, spellings, etc. The only way that they would be useful would be through geocoding the crime data.

If you want to go the geocoding route i you will have to concatenate the address fields (using field calculator) and then use this QGIS plugin if you dont want to pay ESRI:


But, if you have X Y values for the crime data, you can use this tool to create a point shapefile:


if you would like to match each crime to a parcel (or vice versa), you can then use a spatial join on the results of either the geocoding or the XY to point:


  • I followed the instructions on this youtube video: youtube.com/watch?v=h7dcA67_4Jw, but about a dozen of the points are located outside the city (where they should not be), and everything is projected slightly higher than it should be, even when I project them and the parcelbase to the same projection (NAD_1927_US_Feet)
    – T. Kerr
    Apr 24, 2017 at 21:45
  • I'll try the XY point shapefile advice and the spatial join, thanks.
    – T. Kerr
    Apr 24, 2017 at 21:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.