I may be missing something here, but it seems like getting usable US Census data into a shapefile or feature class is much harder than it used to be. Downloading the Tiger block group shapefiles is no problem, but getting and joining the data is. It used to be you could just drill down to the table and area you wanted, download, alter an ID field so you could join data, and you'd be done. Now it seems hideously complicated. As a result, I downloaded the Tiger gdb that the census says is
Ok, block groups are there, along with a variety of tables with various demographic data. They are not prejoined, but whatever, I can still join them by ID. But the field names in the tables are some kind of census code, e.g. B01001e2.
There is a table that gives a description of what these mean, but how do I translate that into field names for the other tables?
There are relationship classes for the tables, but they don't appear to be actually set.