Here’s what we originally thought would be a fun map to make — median household income in various Humboldt County neighborhoods, compared with one another, and as estimated by the Census Bureau’s most recent five-year averages from the American Community Survey.

Check it out below. Go ahead and scroll around and zoom in and such. Hover over or click a census track to get the numbers for that region.  

… and then, probably, you can go ahead and forget you ever saw it, because the data is kind of crap.

Why? Well, when you were hovering and clicking a second ago, did you see the figures for the margin of error in these estimates? Some are better than others, but overall they are not great.

Take the Hoopa square, for example — estimated median household income of $54,306, but with a 36.6% margin of error, meaning that the actual median household income in that area could be anywhere between $34,430 and $74,182. Pretty wide target!

And that’s not even taking into account the American Community Survey’s usual 90 percent confidence interval — meaning the Census Bureau is only 90 percent certain that Hoopa’s median household income is in that range.

These wide margins of error are always a problem the more you break up a dataset into smaller and smaller subsets, and it appears that census tract-level data is small enough to make it all but useless, here in Humboldt County.

Oh well. Maybe some community planner type out there, or someone generally more versed in statistical analysis than I am, can tell me why all this fun tract-level data in the ACS isn’t a total bust for local purposes. Until then, I am declaring it so.