It's been a couple of days since I posted on the New Hampshire recount. At the time, I fully expected that I wouldn't do another post on the topic, but a couple of things that have happened since then changed my mind. First, Scibling Chris Chatham included me in a list of people who he thinks should get off their "soapboxes", stop "hurting America", and focus on the statistical anomaly he's identified. Second, and far more importantly, preliminary recount results are in from a number of precincts.
First, let's look at this "Diebold Effect" thing again. When I took my first look at the results, I decided that a detailed statistical analysis would not be appropriate, and I stand by that. My decision to avoid the potential pitfalls of an inappropriate statistical analysis is not merely because I "assume" that demographic factors don't always explain all of an election result, but because I think that there's a factor that should explain at least some of the results, but which isn't included in analyses that focus on the demographics: campaign effort. We do not have data about where the campaigns chose to focus their efforts during the period between Obama's win in Iowa and the New Hampshire primary.
In fairness to Chris, he did attempt to take this into account after I pointed this out the first time. He found a list of Clinton campaign offices, included their presence as a variable in his analysis, and found that the "Diebold Effect" was still significant. Unfortunately, I don't think that the variable he used as a proxy for campaign effort was remotely adequate. The presence of a campaign office might reflect decisions about where to focus effort that were made early in the campaign, but it's unlikely to reflect last-minute decisions about where to focus effort. (For example, did the campaign shift volunteers from office to office?)
Chris also chides me for not addressing a peer-reviewed paper that he cited which used statistical methods to examine similar accusations that were made in 2004. I've looked at the paper, but I don't think it's an apt comparison. First of all, although Chris claimed that the paper looked at the 2004 presidential primary in New Hampshire, it did not. It actually looked at the general election results there. There are major differences between the primary and the general elections in New Hampshire. The state, as we all know, holds its primary very early in the election season, and the results of the primary have an enormous amount of influence on how the media perceives the campaign for the rest of the primaries. In the general election, on the other hand, New Hampshire has a total of 4 electoral votes, and is not normally the focus of extensive campaign efforts. In addition, the paper in question did not focus on the demographic factors that Chris was looking at. Instead, it primarily looked at prior election data - something that is much easier to do in two-party general elections than it is in party primaries.
That having been said, I am happy that there is a partial recount going on right now. Like Mark, I think it's always a good idea to audit the ballots and make sure that there aren't substantial and systematic problems. Fortunately, that's happening right now in New Hampshire - and some of the precincts in one of the counties that the Kucinich camp was most concerned about have already been recounted.
I took a quick look at some of the results (excel), and was far from surprised by what I saw. In one precinct (Manchester Ward 5), there was a relatively large discrepancy between the machine count and the recounted ballots. In that precinct, the machines messed up and counted vice-presidential write-in choices as presidential votes. As a result, the machines gave more votes to all three of the major candidates than they actually received. Clinton lost 64 votes in the recount there, Obama lost 39, and Edwards 38. (Percentage-wise, Edwards had the largest decline there.) In the rest of the state, the results were entirely consistent with what was found during the 2004 partial recount discussed in the paper cited above: the machines tend to miss some votes, but not very many. As a result, all three major candidates have seen increases in the total number of votes that they received, but the increases are all small, both in terms of the number of votes (excluding Manchester 5, Clinton has received more votes from the recount than the other candidates, with a gain of 52) and in terms of percentage changes (under 1/2 of 1%).
It's entirely possible that statistical analyses will continue to show a significant "Diebold Effect" of some sort, and it might be interesting to try and find the cause. If the balance of the recount matches what was shown, though, I don't think that the "Diebold Effect" is particularly important, and I don't think it's extremely urgent to find the cause - particularly given the difficulties in quantifying things like the last-minute push and campaign strategies.
I am concerned about the habit that these machines seem to have of missing votes all together, but I'm going to hold off judgement on that until I have a better sense as to whether the accuracy of the machine counts is worse than the accuracy in hand-counted areas.
- Log in to post comments
Mike,
One thing has always puzzled me about all this. Why does the US make such wide use of voting machines ? Here in the UK our elections are done using a ballot box, ballot paper with the list of candidates and a pencil and the traditional X in the box of the candidate you want to vote for. It seems to work well. There is little argument over the accuracy of the counts: When there is a close result there is an automatic recount. Most of the counters are bank tellers if I recall. The system also seems to be pretty quick, at a general election enough results are known within about 6 hours of the polls closing at 10pm for the winning party to be known, except when there is a either a hung parliament or a very small majority. Even is the latter cases the result is known within 24 hours max.
Matt,
Take your pick:
1) Laziness
2) Technology fetishism
3) Financial kickbacks from voting machine manufacturers
5) Some combination of the above
Also, I don't know much about the financial costs of hand counting vs. machine counting, but I imagine that the decision makers in areas with machine counting might be swayed by arguments that machine counting costs less.
There are several reasons why a voting district might use machine counting. Whether or not they are good reasons is up for debate.
-JM
Mike, I think you misunderstand why I chide you for not considering the Electoral Studies paper: they have no apparent problem with using statistics to examine vote fraud, they do not include "campaigning effort" (as I recall), and yet they are still able to show an absence of voting technology bias. Why shouldn't the same analysis be used here? Is there something fundamentally different about primaries that makes campaigning effort more important than in the general election? That is an unproven hypothesis which you seem to assume to be true.
FWIW I also tried contacting Clinton's NH offices to get an estimate of the number of staffers, which would give us more sensitivity in that measure, but no one answered the phone.
Because it's not the same analysis. The Electoral Studies paper was able to use past voting behavior in the precincts in question for several elections to show that the votes Kerry received in those precincts were best predicted by looking at the vote totals that other Democrats received there in other elections.
That type of analysis works because the same two political parties are involved in all of the elections studied there. Nader notwithstanding, there are some fairly unsubtle differences between the two parties.
Within-party, things are different. There are no pre-set, clear partisan divisions that last from year to year. The differences between most of the major candidates in the current primary are not large, and in general the candidates agree on many more things than they disagree on. Under the circumstances, I would be shocked if it's as easy to predict who an area will vote for based on demographic factors as it is during general elections.
I agree that it's unproven, but I think that there are very good reasons to think that campaign effort will be more important in the primary (particularly in New Hampshire) than in the general.
To begin with, and as I pointed out above, the choice that voters face in the primary is typically more subtle than the choices people make in the general election. Pick an issue - almost any issue - and compare the positions of each of the three leading Democratic candidates. You'll be hard pressed to find a difference that approaches the magnitude of the difference between the Democrat and any of the leading Republicans. That's going to change things a lot.
Now, let's look at New Hampshire's unique place in the primaries. They have the first primary, and they're the second contest in the primary race. Whoever wins the New Hampshire primary gets a fairly small number of convention delegates, but they get an enormous buzz. The New Hampshire winner becomes the person to beat. They usually get a bump in the polls in other states, and they frequently are able to do a little better in fundraising over the short term.
This means that the candidates give New Hampshire a huge amount of very personal attention before the primary. They're not just spending time doing big media events. They're out talking with people, shaking hands, and conducting retail-level politics.
In the general election, whoever wins New Hampshire picks up four electoral college votes. That's not a lot, particularly when you compare it to battleground states like Florida (27), Ohio (20), or Michigan (17). As a result, New Hampshire is not typically the focus of massive campaign efforts during the general election.
I completely agree that I haven't proven that campaign effort is more important in the primary. Hell, I'm not sure how I'd even go about proving something like that. But I think that I've made a reasonable case for my position, and I haven't heard anyone present any reason to think that it wouldn't be important.
It's entirely possible that statistical analyses will continue to show a significant "Diebold Effect" of some sort, and it might be interesting to try and find the cause.
This seems to be a very cavalier attitude towards our system of voting. I'd rather say we should all be very concerned if there is any indication at all that these machines are having a negative impact on the fairness of our elections.
You seem to imply that is is fine to use any system (even ones designed by commercial and private interests, in secret), so long as it can't be shown with high confidence that the system is broken.
Playing with your spreadsheet, I see several precincts with error rates between 0.5% and 1%: Brookline, Manchester-3, Manchester-4, Manchester-6, Manchester-10, and Manchester-12. I calculated this by adding absolute values of the difference columns and comparing to the total votes recorded for the three candidates on the recount. The reason for looking at that range is that errors in that range are larger than the typical automatic recount threshold of 0.5% but less than the 1% error rate I have heard quoted regarding these machines. It definitely looks like there is a problem in the City of Manchester, and the Town of Brookline should also be worried. Whether problems are widespread elsewhere cannot be determined from this data set.
I notice that all of these precincts are in eastern Hillsborough County. Is that a coincidence, a selection effect on your part, or a selection effect on the part of the NH Secretary of State?
Eric:
It's a selection effect on the part of the Kucinich campaign, which is funding the recount. He's only raised enough funds to cover a partial recount, so he asked the Secretary of State to start with two counties that he found particularly concerning. Hillsborough is one of the two (I don't remember the other off the top of my head).
There has also been an on going discussion about the statistical analysis and I'd say it is unclear how to best analyze these data and accurately estimate the "Diebold" effect. Any statistical minded folks feel free to chime in.
It will be interesting to see what all of the recount data, even if it is limited, look like relative to the original count. I don't think there will be a big discrepancy.