Joining data

Placemark's ability to join data allows you to put new datasets on the map, even if they don't have coordinates. To elaborate, let's say you want to show a map of US States with their average household incomes. It's reasonably simple to find geospatial data with the shapes of the US States by searching for "us states geojson", and possible to find a CSV file of state demographic data, including income. But it's pretty hard to find an existing map data file that has both the shape of the states and the income data as properties.

In this way, you can often find geospatial data and tabular data that corresponds to the shapes in the geospatial data. What's missing is the ability to combine the two. That's what joining does.

Requirements

To use this technique, you'll need two datasets: a geospatial dataset, like a GeoJSON file, or any of the other files supported with importing, and a non-geospatial file in CSV or Excel formats.

And very importantly, they need a matching column. For example, a US States dataset might have state abbreviations like "NY" and "CA", or a countries dataset would have ISO alpha-3 codes for countries, like "USA" and "CHN". What's essential is that this column matches exactly between features on the map and in the dataset that you're joining to the map.

How to join

Okay, so you've got the files. Now you can join data this way:

1. First, import the geospatial file to the map as you would normally.
2. Then start importing the CSV of the non-geospatial data.
3. Select "Join to geodata" under the "Kind" option of the Import dialog.
4. Select the columns to join against, both in the file to import and the map features.
5. Click import.

Tutorial

Want to follow this tutorial exactly? You can download the files I used from this page (click "Download ZIP" in the top right).

What joining does

Joining matches rows in a CSV or XLS file to features on the map, by matching a specific column. When a row in the imported file matches a map feature, the data from that row is joined to the map feature: the data is merged with the feature's properties. If multiple map features match the same row, all of the map features will be joined. For example, if you're joining to countries based on the state's name, and an island nation is represented as multiple distinct Polygon geometries in multiple features (instead of being represented as a single MultiPolygon), then all of the matching features will receive the new properties.

When a row doesn't match any map feature, the join operation will report this and list the non-matching rows.

Joined properties that have the same name as existing properties will overwrite the existing properties.