Disclosure Data Weblog

We're working to provide disclosure of submissions for independent expenditures and electioneering communications we'll receive during the 2010 campaign that's quicker and easier to use. We have some challenges in working with this data so we're interested in your thoughts.

Individuals, groups, and - as of this year -- corporations and unions will be making submissions within 48 hours of activities (beyond a $10,000 threshold) that expressly advocate the election or defeat of federal candidates during most of this year, and within 24 hours if they spend $1,000 just before elections. These are called "independent expenditures" and forms for disclosing them can be found here. Groups, individuals, corporations and unions may also be reporting "electioneering communications" (these are broadcast ads within 30 days of a primary or 60 days of the general election that simply refer to federal candidates).  These must be disclosed within 48 hours of the airing of the ads and the form for reporting them is here.

One of the biggest challenges here is that some of the filings will be amendments intended to add or change information already disclosed, and it isn't always easy to isolate these amendments.  The duplication stems in part from the provision in the law that allows groups to submit these reports in several different ways (e.g. fax, email attachments as well as our web reporting tool and normal electronic filing).

We want to create a process in which users can search for activity based on the race (e.g. California Senate) or a specific candidate as well as the individual or group making the expenditures.  One approach would be to present all of the information from all of the submissions even though some may be duplicated, but that would mean you could not easily compute totals or summaries. 

Another approach would be to delay the availability of searchable data from these submissions until staff here could evaluate them so that only the most recent version of any duplicate reporting would be included in the dataset.  This would allow us to create totals and summaries, but it would probably take a few days so the full set of data would lose some of its real time advantages.

Which of these do you think would work best?  Are there other alternatives we should look at that might solve this problem?  Let us know by using the comments section below.

Thanks

Permalink  |  Comments[4]