I saw this map in a Washington Post article a couple of weeks ago about how much you need to earn to buy a house in different cities around the country. The map comes from an article at HSH.com, which is “the nation’s largest publisher of mortgage and consumer loan information.”

There are pins for each of 27 cities around the U.S. with the salary you would need to buy a median-priced home. In short, this is not a very useful map. Not only does it not visually demonstrate any geographic patterns, but the data markers aren’t sized by the data. This is basically a table placed on a map, but it’s harder to read because your eyes have to jump around the country.

Look, I understand that people like maps—they’re familiar and easy to read—but placing equally-sized pins on a map like this isn’t actually visualizing data, it’s just a more complicated table. What’s interesting, is that just below the map, HSH includes a table that contains the map data plus a few more variables. Why not make the table a bit more visual? Make it a bar chart or another visual so the reader can glean some details. And then provide the data in a downloadable format.


Anyways, using the HSH data, I recreated the map using sized-bubbles, a bar chart, and a scatterplot in the Tableau dashboard shown below (and at Tableau Public here). It’s not fancy, but the data on the map are now visualized and the bar chart makes it much easier to see the differences in salaries across these cities. What I think the two other charts makes clear is how much San Francisco is an outlier relative to the rest of the country.

What do you think? Are there other ways you might present these data? You can download the data at the HSH site and submit your versions in the comment boxes below or on Twitter using the hashtag #PolicyVizRemake.