First Person Singular: Steve Weir
Data is a useful tool for elections officials
By Steve Weir
Contra-Costa County, Calif. Clerk-Recorder
There are two major observations that I have had during my 24 years as County Clerk-Recorder.
First, the people who work in elections are extremely dedicated and ethical. Second, we have in our hands access to a wealth of data that we should use to tell our story. However, many of us miss the opportunity to review and to “own” our data.
I slowly found out in my early days as Clerk, that our elections information management system had TONS of reports on virtually every aspect of our operations.
From simple over-under reports (that can identify individual precinct problems) to rejected vote-by-mail ballots, patterns of problems could be easily identified and tracked.
In 1996, we had a close contest for a California State Senate seat. Out of about 300,000 votes cast, the spread was about 700 votes, not close. However, the losing party asked for a recount. After 25,000 ballots were hand counted, the spread had hardly changed and the recount was called off.
As part of this process, I noticed that 3,200 vote-by-mail ballots had been rejected, almost 4 percent of the total vote-by-mail ballots cast. Most of these arrived after election day. No one seemed bothered by this statistic. No one except me. These were voters who did not have their ballots counted.
We developed an educational program and watched these numbers decline significantly. This felt right. And, with this base line, we were able to discover other occurrences of rejections.
In the March 2, 2004 Presidential Primary, we had a spike in rejections due to late arrival. We drilled down and found that the “problem” was in our mandatory vote-by-mail group. For mandatory vote-by-mail, we pay the return postage through a business reply category.
While our late category of rejection was under 1 percent we had a spike and quickly were able to see that it was in this category. Come to find out, the clerk who processed this business reply mail was out on the Monday and Tuesday of election week. So, these voters had almost 12 percent of their mail rejected for being late.
This prompted us to meet with our Post Office representatives to guarantee they would process this mail in a timely manner. Had we not tracked this data, the problem could have continued over time with no one being any-the-wiser.
As we also tracked rejection rates for “no-signature match”, we witnessed a spike in rejections in June and November, 2010. With this on our radar, we ran a program equating age and rejection. We thought that our older voters might be responsible for the spike in rejections.
Quite to the contrary, it worked out that our voters (many of whom had been part of the 2008 registration efforts) who were in the 20 to 29 age group, accounted for 6.2 percent of our total vote-by-mail ballots cast while they accounted for 30 percent of the total rejection rate for no-signature match.
We designed an education program to try to address this phenomenon. It seems that this “new” generation is not use to using their signature as a security device. We’re hoping that an educational effort via social media can help address this issue.
All of us have been beseeched with requests from lawyers, academicians, activists, etc. only to have that data recast and thrown back at us. I believe that many of us a “gun shy” when it comes to data. My position is that we ought to “own” our data, its right at our finger tips and available for us to track and to adjust our operations when warranted.
If you are interested in seeing a copy of our report, go to: http://www.cocovote.us/
In the right column (second selection from the top) is the election report. Click on it to download.
(Editor’s Note: Do you have a story you’d like to share with your fellow elections officials, academics, vendors and geeks? Please drop us a line and let us know, we’d love to hear from you