For the past year we’ve been busy adding data and features to Influence Explorer, our central source for data on money and influence in politics. From our original three data sources (federal campaign finance, state campaign finance and federal lobbying) we’ve now expanded the site to include earmarks, federal spending, contractor misconduct and EPA enforcement data. We’ve also released tools such as Checking Influence and Inbox Influence that put political influence data in a context that’s relevant to users.
After tackling these larger projects we decided to step back and revisit a few things that we didn’t get perfect the first time around. In a number of places we’ve tweaked our methodology, cleaned up our data and added more context. The result should be more accurate and useful information across the board. Here’s a rundown of what’s changed:
- Subsidiary organizations are now handled more intelligently. There are fewer erroneously listed subsidiaries, and the amounts listed on the Top Donors table will more often match the amounts listed on the Top Contributors table.
- Generic industries and occupations, like “unknown” and “retired” are no longer shown on politician donor tables. Instead there’s a new info line saying how much of a politician’s donations came from identifiable sources.
- The Wikipedia bios have been hand reviewed for accuracy. For example, the construction equipment company Caterpillar is no longer described as “the larval form of a member of the order Lepidoptera.”
- When a state politician later runs for federal office we now show a link between the two pages.
- The rules for what contributions to include have been tweaked. The most noticeable change is that candidate self contributions are now included, which makes a big difference in cases like Meg Whitman or Jon Corzine.
- The data listed in the federal spending section has been made much more accurate. Previously we allowed very loose matches to be displayed; now we require an exact match on the company name.
We hope that these changes will lead to a more complete picture of the data and fewer head-scratching moments. And be on the lookout for a whole slew of new data releases in the next month.