When is an Election Over? Depends Who You Ask – and Why


[Image courtesy of WithoutWax.TV]

Not surprisingly, the recent news about the muddled finish in the GOP Iowa caucuses has got people talking about what it means to say an election is “over”.

To an election official, an election is over when the outcome has been certified according to applicable state or local law. At that point, the process for that election is “final-final” and in the books. That’s why election administrators are so insistent about calling Election Night returns “unofficial” returns; experience teaches that lots of factors – including everything from math errors to multiparty litigation – can make the Election Night results turn out to be incomplete or incorrect.

To candidates, campaigns and the media, however, an election is usually “over” on Election Night. At that point, the running narrative – horse race, policy debate, or something in between – is generally seen to be resolved and peoples’ attention moves on to next steps: how a candidate will handle victory/defeat, what the outcome means for one or more running policy debates – even who will (and won’t) be a candidate in the NEXT election.

The challenge, of course, is to find a way to bring these two views into closer harmony. Just about every election cycle, one or more administrators gets into hot water because Election Night returns weren’t made available fast enough – starving an audience hungry for news. When that happens, I’m generally sympathetic to those who say we needn’t rush the election process and that everyone concerned needs to be more patient about waiting for the official outcome.

At the same time, I do think the field of election administration needs to think about ways in which we can shorten the amount of time it takes to go from unofficial returns to certified results. That doesn’t mean rush the process – which needs time to complete – but rather to treat timeliness of final results (including any necessary cross-checks or audits) as a necessity rather than a luxury. How to balance this with other priorities is of course an open question, but in the current environment states and localities may want to look for ways to finish their counts before their statutory or traditional deadline for doing so.

If nothing else, shortening the window between unofficial returns and final certification would eliminate the “are we there yet?” phenomenon that always seems to result when a storyline born on Election Night begins to collide with actual election results.

5 Comments on "When is an Election Over? Depends Who You Ask – and Why"

  1. Matt Masterson | January 20, 2012 at 12:25 pm | Reply

    This is a great post and one that falls near and dear to all election officials hearts. I have a lot of thoughts on this but the one that comes to mind first is the value, timing, and effectiveness of audits. In a previous post there was a little discussion of an academic paper regarding moving towards a system that favors strong audits and moves away from certification. We cannot have any version of that and shorten the time period as suggested in this post. To do an audit well takes time. So which one is it?

  2. Ultimately it’ll have to be both … that’s the “how to balance this with other priorities” part – but if we get to the point where auditability is “baked in” voting technology – which is a goal suggested by the Stark/Wagner piece I blogged about last week – it may be possible to compress the amount of time it takes to go from a “pile of ballots” to audited, certified returns.

  3. Good post. Risk-limiting audits bring a very helpful property to this discussion. The point of a risk-limiting audit is getting the outcome right, not necessarily getting the exact count right. In a contest with a wide margin on election night, it usually doesn’t take much of an audit to come to the conclusion that the apparent winner did indeed win. So an abbreviated canvass and audit could be done quickly and the outcome could then be announced early before all the rest of the votes (including provisionals, which take a while), and the detailed canvassing and other paperwork is done to come up with the final official count. On the other hand, if the margin is really narrow, the appropriate message is “the outcome appears really close and we will be looking very carefully at all aspects of the election and doing a good canvass and audit before we declare a winner”. That may take time, and people should understand that.

    I don’t know if it makes sense to adapt the procedures to the apparent margin like that, but if the possibility of a quick authoritative answer is important, an RLA like this could fit the bill.

    In the case of contests like the Iowa caucuses, just knowing that the race was very close may be most of what the pundits and public really need to know right away anyway.
    On the other hand, for contests like caucuses in which the outcome is really the number of delegates to be awarded, the actual margin may be very small, and if people need to know the actual numbers, it may take a while to get confidence about the result.

  4. For those interested in election integrity, and passionate members of the public, some elections are never complete, as long as doubts remain with the credibility of the results, along with holes in the prevention of similar future problems and doubts. Much of the current incomplete work for integrity was initiated by the 2000 and 2004 elections. There remain important lessons from the election of 1876 and a variety of recent debacles such as Sarasota in 2006.

  5. There is a catch phrase frame for this topic- “election finality”. This finality is seen as an advantage by election officials- the end of the period during which the procedures taken as well as the results obtained are under scrutiny and subject to criticism.

    We are on the cusp of an era in which the public at large may finally have an opportunity to interpret and tabulate ballots and check official interpretations of each ballot thanks to scanner and internet technologies. Note that these technologies merely enhace access to the a voter verified paper record of voter intent…I’m not talking about internet voting or DRE.

    Access to copies of ballots and their interpretations could have some impact on deadlines for certification. In Colorado we see the county clerks right now pushing to prevent any citizen access to ballots during the canvass phase of the election, but its obvious that one wants all the criticism of an election to arrive at a point at which it can be digested and incorporated in the determination of results and outcome. The “elective” or “targeted” audit unit approach of the risk limiting audit gives us just what is desired- an orderly way to input citizen criticism into an election canvass- by adding ballots to the randomly selected units for audit. This obviously works most efficiently with a ballot level audit.

    Holding back the public access to ballots till after certification has the effect of creating a potential storm of controversy after it is too late to make a remedy until the next election.

    Election finality has up till now been guaranteed by destruction of records. In Colorado, our antiquated municipal law for elections requires destruction of ballots 6 months from the election. With digital storage, it should be possible to keep copies of election records virtually forever. Will the “election finality” lobby prevent us from doing that? Its something that remains to be seen.

    All of the above rests upon the presumption that ballots are fundamentally anonymous, and any exceptions are handled by duplication, redaction or witholding.

Leave a comment

Your email address will not be published.