What does “election hacking” mean to the public?

Yesterday I wrote about a recent poll I conducted that revisited the question of whether voters thought computer hacking was a major problem in the November 2016 election.  That post noted that more Americans have come to believe computer hacking was a major problem in 2016 than they believed back then.  However, the majority of opinion change has come from Democrats.  You can read that post here.

The question that was asked in the survey specifically mentioned “computer hacking” in the “administration of the election.”  However, the issue of “hacking the election” rarely is that specific when it comes up in the news or informal conversations.  So, I decided to ask the respondents this question at the very top of the survey:

There has been talk in the news recently about computer hacking in American elections. When someone talks about hacking American elections as a general matter, which of the following do you think about first?

The following table shows the possible response categories and how they were distributed among the respondents.  A plurality of respondents chose one of the two options involving foreign actors, whether using social media to influence voters (20%) or breaking into the computers that run elections (25%).  A good number of people, 20%, said that “nothing in particular” came to mind when they hear talk of computer hacking in elections.

Table 1.  Question: When someone talks about hacking American elections as a general matter, which of the following do you think about first?
All respondents Democrats Republicans
Foreign actors using social media, like Facebook, to influence how people vote 20% 28% 18%
Americans using social media, like Facebook, to influence how people vote 9% 7% 12%
Foreign actors trying to break into computer equipment used to run elections, like voter databases and voting machines. 25% 33% 21%
Americans trying to break into computer equipment used to run elections, like vo
ter databases and voting machines.
17% 12% 24%
Something else 8% 7% 8%
Nothing in particular 20% 13% 18%
N 2,000 880 603

There is a partisan divide in how respondents think about the topic of election computer hacking, but the pattern is more complicated than Democrats simply thinking there’s a big problem and Republicans not. A majority of Democrats chose one of the two responses that focus on foreigners, compared to only 39% of Republicans.  However, Republicans were much more likely to choose the response about Americans breaking into election equipment.  Like I said, the partisan divide on this question is not straightforward.

It’s hard to know what’s going on here, with only one question in a limited survey.  However, my favorite hypothesis is that this is evidence that Democrats are focused on the “Russian hacking” narrative about the 2016 election, whereas Republicans, when they think about problems associated with the election, are drawn toward corruption of election administration itself.  My guess is that had I asked respondents which was the bigger election administration problem, breaking into voting equipment by foreigners or inside corruption of the process, the parties would have neatly divided on the question — but, that’s a question to explore in the future.

Returning to the issue of political knowledge, it’s not surprising that people who follow the news most closely are more likely to have an opinion about the question (in other words, the “nothing in particular” response is less common) and that the partisan patterns seen above are heightened.  This is shown in the next table

Table 2.  Question: When someone talks about hacking American elections as a general matter, which of the following do you think about first? (Sub-sample:  respondents who report following the news “most of the time.”
All respondents Democrats Republicans
Foreign actors using social media, like Facebook, to influence how people vote 30% 37% 25%
Americans using social media, like Facebook, to influence how people vote 6% 4% 9%
Foreign actors trying to break into computer equipment used to run elections, like voter databases and voting machines. 32% 41% 22%
Americans trying to break into computer equipment used to run elections, like vote

r databases and voting machines.

16% 8% 24%
Something else 7% 6% 8%
Nothing in particular 9% 5% 11%
N 923 465 603

In particular, three-quarters of high-information Democratic respondents chose one of the two foreign-actor responses, whereas high-information Republicans seem to have a variety of first-thoughts about election computer hacking.  (In fact, it’s as if the high-information Republican respondents are choosing the response categories almost randomly, in stark contrast with the Democrats.)

To put yesterday’s and today’s posts together, it is interesting to note that there is a link between first thoughts concerning election hacking and the degree to which one thinks that computer hacking was a major problem in 2016.  Respondents saying that foreign hacking of computers used in election administration was the first thing that came to their minds were the most likely to say that computer hacking was a major problem in 2016.  (See the figure below.* Click on the figure to enlargify.)  This was especially true among Democrats, and only somewhat true among Republicans.

I concluded yesterday’s post by suggesting that Democratic and Republican constituents were likely to exert different levels of pressure on their parties’ legislators to do something about computer security in elections.  Today’s post suggests an amendment to that conclusion. In particular, the Democratic mass public seems convinced that (1) computer security in elections is a big problem, (2) the problem comes from outside the country, but (3) they can’t choose whether social media manipulation or voting machine hacking is a bigger problem.  In other words, security is a problem, and we’re being attacked from abroad.

The Republican mass public is not convinced that computer hacking of elections is a major problem; to the degree it might be a problem, they are more conflicted over whether it’s a domestic or foreign threat.

As is often the case in politics, it’s the side that has a clear diagnosis of a problem and its solution that drives the debate.  In that case, it’s the Democrats.  Of course, they don’t have the majority, at least in the nation’s capital, which may be a prescription for a lot of talk, and not a lot of action, from our national legislators.

*The figure originally had an error in how the x-axis categories were labeled.  It has been corrected.

Partisans Divide over Election Hacking

A recent survey of 2,000 adults shows that Americans have become more concerned about election hacking than they were in 2016, and that a partisan divide has widened over these concerns.

This is the second in a series of surveys I’ve taken in the past several months, where I have looked at opinions held by Americans about problems facing the electoral process. In a series of previous posts, I looked at public attitudes toward the Pence-Kobach Commission, on the heels of its termination in January. (You can find those posts here, here, here, and here.)

In this post, I look at the issue of hacking.

The story starts in November 2016, when I threw two questions onto the end of the Survey of the Performance of American Elections. These questions asked respondents to report how much of a problem they though computer hacking was in the administration of elections in 2016, both nationwide and locally.

Recall that news and rumors of hacking — of social media, campaign websites, voting machines, and voter registration files — were a part of the news diet at the time, but it hadn’t developed into the major, multi-pronged story that it is now.

Back in November 2016, 17% of respondents thought computer hacking in elections was a major problem nationwide, while 10% thought it was a major problem locally.

What a difference a year makes. Last week, when I asked identical questions again, the percentage of Americans believing computer hacking in 2016 was a major problem had doubled — to 38% who believed it was a major problem nationwide, 20% locally.

Table 1.  Question:  How much of a problem do you believe computer hacking was [nationwide/locally] in the administration of elections in 2016?

Nationwide

Locally

Nov. 2016

Mar. 2018 Nov. 2016

Mar. 2018

Major problem

17%

38% 10%

20%

Minor problem

29%

28% 19%

25%

Not a problem

28%

15% 46%

29%

Not sure

26%

20% 25%

26%

N

10,199

2,000 10,199

2,000

What’s especially noteworthy about this change is the partisan detail. Although respondents in all three major partisan categories (Democrats, Republicans, and Independents) were more likely to view computer hacking in 2016 as a major problem, the biggest shift came among Democrats, who went from 23% viewing hacking as a major nationwide problem when asked about it in November 2016, to 56% when asked the same question this month. (See the accompanying figures; click on any of the figures to enlarge them.) The fraction of Independents viewing hacking in the 2016 election as a major nationwide problem grew from 16% to 30%; the fraction among Republicans grew from 10% to 18%.

Similar partisan patterns appear when we look at the question of computer hacking as a local election administration problem. Among Democrats, the percentage saying that computer hacking was a major local problem in the 2016 election was 14% in November 2016, compared to 29% when the same question was asked this month. Among Republicans, the percentage had grown from 4% to 9%; among Independents, it had grown from 10% to 17%.

These results have important implications for the politics of election hacking and the policy response. Here are two quick thoughts:

  • Leaving aside partisanship, there is greater concern with hacking as a nationwide problem than as a local problem. This may mean greater pressure on state and national officials to address problems of election cybersecurity than on local officials. Of course, as anyone in the elections business knows, the decentralized nature of election administration in America means there are lots of small jurisdictions that are probably the most vulnerable to attacks. Whether political pressure will line up with the nature of the threat is a question that is raised by these results.
  • Adding partisanship to the mix, and there is a significant mismatch between the Democratic and Republican mass publics about the severity of the problem. To the degree that election security reaches the political branches, this means that Democrats are likely to feel more pressure by their strongest supporters to do something about security threats, such as pass legislation like the bipartisan Secure Elections Act, than Republicans. Luckily, state and local election administrators don’t need partisan pressure to be attentive to issues of security, but the partisan perception of the threat may make it hard for them to get much help legislators on the issue, depending on local circumstances.

That’s enough on the partisanship angle for now. The survey contained a couple of other questions about perceptions of the threat of computer hacking in elections, which I hope to write about in the coming days.

Methodological note. The March 2018 survey referenced in this post was conducted by YouGov as a part of their omnibus survey. The November 2016 Survey of the Performance of American Elections was also conducted by YouGov as a special project. Both surveys were weighted to produce a representative sample of American adults. The questions about computer hacking asked in each survey were identical.

A Viewer’s Guide to Special Election Watch

Today starts a regular feature of the MIT Election Data and Science Lab called “Special Election Watch.”  The idea is to follow special elections in 2018 as a guide to the extent of the Democratic swing in the November 2018 election.  First, a couple of words of background, and then a guide to the graph we will be updating regularly.

Political scientists have long followed the partisan “swing” from one election to the other.  Analysis of the “uniform national swing” has been a staple of discussing British politics going back to the 1940s.  The question of whether the swing is so uniform in the U.S. has a distinguished pedigree, perhaps most notably represented by a small book by Tom Mann, Unsafe at Any Margin, in 1979 and an article by Gary Jacobson,  “The Marginals Never Vanished” in 1988.

Despite the fact that the partisan swing of electoral fortunes varies across districts in a particular election, the average swing across districts is a good starting point for gauging the degree to which one of the political parties will be favored.

Which brings us to the special election watch.  It’s widely accepted that 2018 will be a good electoral year for the Democrats.  But, by how much?  One common measure is the so-called generic congressional poll.  Another way to measure relative electoral strength is to calculate the change in electoral fortunes of the parties as they defend legislative seats in special elections.  That’s the approach we’ll be following here.

Ballotpedia’s 2018 election calendar currently lists over 40 special elections between now and June for state legislative seats.  As each one is held, we’ll be following the returns and comparing them to the results when the seat was last contested, usually in 2016.  The accompanying graph shows the results for this year’s special elections up to last Tuesday. (Click on the graph to biggify it.)

Here’s the guide to how to read the graph.  Each row represents a state legislative special election, with the district indicated on the left and the date of the election on the right.  The code for the district is the postal state code + “H” for House or “S” for Senate + district number.  The circle represents the percentage of the vote received by the Republican candidate in the last regular election; it’s red if the Republican won and blue if the Democrat won.  The arrow points to the Republican percentage in the special election; the arrow is red if the Republican won the special election and blue if the Democrat won.  At the very bottom, we show the average.

As of last Tuesday, taking all the special elections into account, the average swing has been about 16 points in favor of Democratic candidates.

Of course, there is more to this graph than just the average.  For instance, a large number of these seats were originally won without a contest.  Indeed, of the 15 elections shown here, 8 were originally won without a contest, 6 by Republicans and 2 by Democrats.  Not surprisingly, the average pro-Democratic swing was greatest in the previously uncontested seats (27 points) than in the seats that had been contested (4 points).  About half of all state legislative races were uncontested in 2016. If the patterns in the special elections hold in the general elections, we can expect a sharp drop in uncontested elections at the state level, which will be a significant event in itself.

This past week, there has been considerable press coverage of the special election in the 97th Missouri House district, which flipped to the Democratic column after it had given Donald Trump a large majority in 2016.  While seat-flipping is certainly notable, the larger story was the overall mobilization of Democratic voters in the four Missouri districts that had special elections on the same day.  Two of the districts (the 39th and 144th) had gone uncontested in 2016.  The fact that these districts attracted strong Democratic candidates, one of whom came close to winning, is almost as important as the Democratic victory in the 97th, at least from the perspective of measuring the relative appeal of the two parties’ legislative candidates these days.

The 2018 election year is just starting.  It’s a long way to November.  However, as Gary Jacobson and Sam Kernell taught us in their classic book, Strategy and Choice in Congressional Elections, the most important events of a congressional election year happen at the beginning, when the candidates, parties, and potential backers size up the field.  So far, the elections that are currently being held are consistent with even more good Democratic candidates jumping in and even more good Republicans staying (or moving to) the sidelines.

 

The Public Says Good-Bye to the Trump Vote-Fraud Commission, Part IV: Messaging about Fraud

This is the fourth of a four-part series looking at public attitudes related to President Trump’s fraud commission.  Part I introduced the series and explored whether voters had become more concerned about vote fraud since late 2016.  Part II explored who was knowledgeable about the commission.  Part III examined public opinion about the termination of the commission.  Today, I look at messaging about voter fraud, and then offer up some summary thoughts about the work of the commission, its termination, and the future.

Messaging about voter fraud

The value of a presidential commission is that it draws attention to an issue and can serve as a focal point as public opinion is mobilized for policy change.  Both ardent supporters and opponents of the Presidential Commission on Election Integrity saw the commission as filling this role in the case of vote fraud and voter registration.

Because the most prominent member of the commission, Kansas Secretary of State Kris Kobach, has developed a reputation as a determined opponent of voter fraud, it was reasonable to expect that had the commission continued conducting business, it would have been an amplifier of claims about the high prevalence of voter fraud in the U.S.  The commission never served this function, both because of its premature demise and because the news surrounding the commission was muddied by its controversy.

However, let’s consider the counterfactual.  What if the commission had not been terminated, and what if Kobach’s message about high rates of fraud had been given clear expression?  What would the effects of this messaging be?

We’re helped in answering this question by Kobach, who was interviewed by Breitbart on the subject of voter fraud on the day the commission was terminated.  In that interview, Kobach claimed that the vote fraud commission had revealed:

  • 938 convictions for voter fraud since the year 2000
  • Fewer than 1 in 100 cases ends in a conviction
  • In Kansas, alone, there are 127 known cases of non-citizen aliens registering to vote
  • In 21 states, there were 8,471 cases of double voting discovered

I turned these claims into a question that was asked of my survey respondents. (Go back to the previous posts to see details about the sample and other analysis.)  In particular, I randomly presented one of these claims to each respondent, identifying it as having been made by “the vice chair of the fraud commission.”  I then asked, “Does this statement make you more or less concerned about voter fraud?”

A plurality of respondents (47%) said they were unmoved by the statement they were shown, 40% said it made them more concerned about fraud, and 13% said they were less concerned.  (See Table 1.) Most Democrats said their views were unchanged; most Republicans said they were more concerned.

 

Table 1.  Question:  Does [the statement you were just shown] make you more or less concerned about voter fraud?
All respondents (N = 2000) Democrats (N=851) Republicans (N=652)
More concerned 40% 30% 55%
Less concerned 13% 16% 9%
No change 47% 55% 36%

Thus, the respondents processed claims about voter fraud through the lens of partisanship.  This is consistent with much of the modern research into the effects of persuasive communicates, which generally teaches us that people readily internalize information that confirms prior beliefs and reject information that is dis-confirming.

This process is typically the most potent among the most politically engaged, who are the most aware about “what goes with what,” that is, most aware of the positions associated with the two parties.  And, true enough, that pattern shows up here, as well.  For instance, while 55% of Democrats stated that the statement they read about voter fraud did not change their concerns about fraud, 72% of the “hyper-aware” Democrats said they were unmoved.  (Recall, the “hyper-aware” respondents were those who both (1) closely followed news about the commission and (2) could pick out Kobach as a member of the commission.)  Among hyper-aware Republicans, 61% said they were even more concerned about fraud after reading the statement, compared to 55% of Republicans overall. 

In other words, among the most engaged partisans, being presented with a claim about voter fraud just pushed them further apart from each other.

A similar pattern also emerges when we examine the responses to these statements in light of how concerned the respondents said they were about vote fraud at the start of the survey. (See Figure 1).  Among those who initially said they were the least concerned about vote fraud, 73% said that the statement they read about fraud didn’t change their mind.  Among those who initially identified themselves as being the most concerned about fraud, 72% said the statement made them even more concerned.

As an aside, the pattern in Figure 1 persists among all levels of attention to the commission, or even all levels of attention to politics in general.  What this means is that attitudes about vote fraud are probably already pretty entrenched among citizens, and that claims about fraud have little-to-no influence on those attitudes, at least in the short run.

Among the four statements, which had the greatest effect on concerns about fraud? Table 2 provides the answer to the question.

Table 2.  Effect of statement read to respondent on respondent’s self-report about change in attitudes about vote fraud. (N=1,998)
More concerned Less concerned No change
938 convictions for fraud since 2000 30% 18% 52%
Over 100,000 cases of voter fraud prosecuted since 2000 35% 11% 55%
127 known cases of non-citizen voters in Kansas 47% 11% 42%
8,471 cases of double voting in 21 states 47% 12% 41%

What stands out here is the fact that the last two comments tended to elicit more expressions of additional concern than the first two.  Figuring out why this is requires some speculation, because each of the items mixes and matches stimuli.  Still, it strikes me that the last two items pertain to specific types of voter fraud (non-citizen voting and double voting), whereas the first two items refer to fraud in general terms.  It is possible that specific cases of fraud are more compelling than the general problem of fraud.  Or, it is possible that non-citizen and double voting in particular are especially compelling ways to frame the fraud question.  In any event, I am sure that others who are more expert in the field of persuasive communications than I am will be exploring this question further.

*      *      *

This has been a fast-and-furious tour of the new survey data about the Trump fraud commission and its termination.  I hope to be writing more about this in the weeks and months ahead.  For now, it is important to note that support for the commission and its termination was not as firmly linked to partisan attitudes as attitudes about fraud itself.  I have gone on record as saying that the termination of the commission only shifts the politics of vote fraud to other venues.  It remains to be seen whether those venues will be seen as having greater authority than the commission, and thus (perhaps) greater influence on what the public believes about fraud.

One reassuring message from the survey responses is that even Americans who were the most concerned about voter fraud were not up in arms about the commission’s termination.  This suggests an opening for bipartisan endeavors to secure the voting rolls that may be effective with the broad middle of public opinion.  There are already effective models of such efforts, ranging from states that carefully adhere to the NVRA, to the Electronic Registration Information Center.  Efforts such as these are likely to be more effective in securing voter registration lists while maintaining access to the ballot box than promoting messages that polarize attitudes further than they already are.

The Public Says Good-Bye to the Trump Vote-Fraud Commission, Part III: Opinions about Termination

This is the third of a four-part series looking at public attitudes related to President Trump’s fraud commission.  Part I introduced the series and explored whether voters had become more concerned about vote fraud since late 2016.  Part II explored who was knowledgeable about the commission.  In today’s post, I look at support for the commission’s termination.

Among the respondents who said they had heard of the commission (which as 79% of respondents), 44% agreed with the commission’s termination at least somewhat, 31% disagreed, and 25% had no opinion.  (See Table 1.)  Among the aware, 53% agreed with the termination.  This rises to 63% among the hyper-aware.

Table 1.  Question:  Do you agree or disagree with President Trump’s decision to terminate the election fraud commission?
All respondents (N = 1,568) Democrats (N = 700) Republicans (N=531)
Strongly agree 27% 41% 17%
Somewhat agree 17% 13% 25%
Somewhat disagree 17% 11% 24%
Strongly disagree 14% 18% 8%
Don’t know 25% 17% 26%
No response 0.3% 0.1% 0.0%

The relatively large number of “don’t know” responses is typical, in my experience, when asking about questions related to election administration.  Not surprisingly, Democrats were more likely to agree with the termination (54%) than Republicans (42%) and more likely to have an opinion, as well (17% “don’t know” vs. 26% for Republicans).  Given the large percentage of Republicans who expressed concern over fraud in the first question, the big surprise here is that a plurality of Republicans actually agreed with the commission’s termination — a majority if we exclude the respondents who had no opinion.

Another surprise is the relatively large number of Democrats who strongly disagreed with the commission’s termination. In fact, more Democrats disagreed with the commission’s termination than Republicans.  Part of the reason for this is that Democrats were just more likely to express an opinion, but even among those expressing opinions, Democrats were more likely to disagree.

Who are these Democrats?  First, they were much more likely to say they were very concerned about vote fraud at the outset (45%, vs 8% of all other Democrats).  They were also more likely to say they were moderate (44%) than other Democrats (38%).  They were also less attentive to politics in general and were much less likely to identify Kris Kobach as a commission member (32%) than other Democrats (68%).

Thus, the Democrats who strongly disagreed with the commission’s termination were not especially attentive to the commission’s work and had more ideologically moderate views than the rest of their co-partisans.  In contrast, the Republicans who strongly disagreed with the termination of the commission were not all that different from the rest of their party.

It is quite possible that the Democrats who strongly disagreed with the commission’s termination simply had the wrong idea about what the commission’s charge was.  Unfortunately, the limitations of this survey didn’t allow me to probe that question further.

In the end, the public seemed relatively satisfied to see the fraud commission go away.  Whether they would have had the same opinion had the commission been able to pursue its mission through 2018 is something we will never know.

Tomorrow’s post: The effectiveness of claims about vote fraud.

The Public Says Good-Bye to the Trump Vote-Fraud Commission, Part II: Knowledge of the Commission

This is the second part of a four-part series concerning public attitudes toward the Trump Vote-Fraud Commission.  The first part includes an overview of the series, some information about the survey methodology itself, and a substantive section on attitudes about fraud.  Today, I take on the issue of knowledge about the commission itself.

President Trump’s fraud commission is one of those inside-baseball issues that elicits three types of reactions: (1) strong, informed reaction among the most politically knowledgeable, (2) strong, uninformed reaction among a larger group of engaged political spectators, and (3) a variety of “non-opinions” among the rest.  I used two questions to see how closely the public followed the commission’s work, to see if we could discern differences of opinion among these three groups.

First, I simply asked respondents to report which of a series of statements most closely described how closely they had followed the commission’s work.  Table 1 shows those statements, and the fraction of respondents who chose each.

 

Table 1.  Question:  Which of the following statements most closely describes how much you paid attention to President Trump’s election fraud commission before you took this survey? (N=2,000; 0.06% gave no response)
I really hadn’t heard about it before.

21%

I had heard about it, but I didn’t pay much attention to it.

30%

I had heard about it, and remember reading a little bit about it.

28%

I had heard about it, and followed news about it with interest.

21%

 

Overall, the public appeared to be split about 50/50 between those who paid at least some attention to the commission and those who didn’t.  Democrats and Republicans followed it in roughly similar proportions, 51% for Democrats and 54% for Republicans.

Of course, the nature of questions like this is that people tend to over-report being attentive to the news of the day.  To try and get a sense about how closely people really followed the commission, I also gave the respondents a test, to see if they knew of the commission’s most prominent member, Kris Kobach.

I asked them, “Based on what you might have read or heard about the voter fraud commission, which of the following prominent election officials served as one of its leaders?”  I then gave them four choices, which were shown to respondents in random order.  One choice was Kobach.  The other three were prominent election administrators from around the country, none of whom served on the commission.

Overall, 40% of respondents correctly picked Kobach.  Of those who reported that they had followed news about the commission with interest, 60% got the answer right.  Among everyone else, only 35% got it right, a result barely better than chance.

These results aren’t intended to play “gotcha” with the American public, but rather, to show that it is easy for those of us in the election administration world to over-estimate the attention being paid to things like the fraud commission.

For the rest of this post, I will call the people who reported that they had followed news about the commission with interest “aware” (21% of respondents) and those among this group who correctly picked out Kobach as the “hyper-aware” (13%).

 

Tomorrow’s post:  opinions about termination

The Public Says Good-Bye to the Trump Voter-Fraud Commission: Introduction to a Series

This post begins a four-part series about some public opinion research I recently did to gauge attitudes about the termination of the Presidential Advisory Commission on Election Integrity.  The survey was conducted in late January of a representative sample of adults, sample size of 2,000.  There’s more to be said from this data than can fit into one blog post, which is why I’ve decided to stretch it out across four days.

To whet your appetite for what is to come, here’s the executive summary of the whole series:

  • Respondents were slightly more concerned about voter fraud in January 2018 than immediately before the 2016 election.
  • Respondents were split about 50/50 in whether they followed the work of President Trump’s fraud commission.
  • 44% of respondents agreed with the commission’s termination at least somewhat, 31% disagreed, and 25% had no opinion. A majority of Democrats and a plurality of Republicans approved of its termination.  More knowledgeable respondents were more likely to agree with the termination.
  • Respondents who reported they were already the most concerned about vote fraud report becoming even more concerned about vote fraud when read statements making claims about the frequency of fraud.

Today, I introduce the subject of the series and evidence about overall opinions about vote fraud.  In subsequent days I will touch on knowledge of the commission, options about terminating the commission, and some evidence about messaging concerning vote fraud.

Introduction to the Series

Almost a month ago, President Donald Trump terminated his Presidential Advisory Commission on Election Integrity.  The commission’s request during the summer for states’ voter files set its early life off to a rocky start, and helped create a degree of bipartisan opposition among election administrators that was surprising in its near-unanimity.

Many were worried that the commission would create doubts in the public’s mind about the integrity of the electoral system.  Democrats were especially worried that the commission’s work would lend credibility to discredited claims that millions of illegal votes were cast in the 2016 election.  This could have laid the groundwork for the weakening of voter-registration-protection provisions in the National Voter Registration Act, the “motor voter” law.

We’ll never be able to test whether the greatest fears of the commission’s opponents came to fruition, but we can catch a small glimpse into how the public thought about the commission’s work during its truncated lifetime, and its demise.

To capture a snapshot of public attitudes, I commissioned YouGov to ask a small battery of questions about election fraud and the commission’s demise to 2,000 respondents last week.  (The poll was in the field between January 23 and 26, 2018.  It would have been in the field earlier, but university IRBs move slowly, as they should.)

The questions I asked were intended, first, to see if attention to and worry about voter fraud had changed in the year that has passed since the 2016 election.  I was also interested in whether the public had knowledge about the commission at all, and whether they approved of its termination.  Finally, I wanted to see whether the types of messages the commission may have publicized, had it not been terminated, would have resonated with the public.

There were two major take-aways for me.  The first had to do with the facts.  About half of Americans followed the commission, though a small minority followed it very closely.  The public has become slightly more concerned about voter fraud, compared to the pre-2016 election period, with Democrats showing the most movement.  The public mostly supports the termination of the commission, with Democrats strongly supporting its termination and Republicans being more equivocal.

The second take-away relates to how public opinion works in this partisan era.  For the past decade, as partisanship has started to structure so much of how Americans view politics, the major framing device to understand this phenomenon has been something called “motivated reasoning.”  Applied to opinion about public policy, motivated reasoning helps us understand how citizens interpret new information, including conflicting information, through a partisan lens.  It also helps us understand how people with extreme views just keep getting more extreme.  As we’ll see, when presented with claims about evidence of voter fraud, respondents who were already the most concerned about voter fraud became even more concerned.  The respondents who were the least concerned about fraud out the outset either became less concerned, or didn’t change their stance, when presented with the same claims.

Overall opinions about voter fraud

To start, I asked respondents how concerned they were about voter fraud, along a five-point scale.  The distribution of responses is given in Table 1.

 

Table 1.  Question:  The following are two statements that might be used to describe voting and elections.  Using the scale provided, please indicate which statement more closely describes how you feel. (N = 2,000; 9% gave no response)
I am not concerned about vote fraud

1

2 3 4 5

I am concerned about vote fraud

18% 14% 17% 17%

26%

For simplicity’s sake, let’s call people who rated themselves a 1 or 2 “not concerned” and those who rated themselves a 4 or 5 “concerned.”  Among the respondents, 43% said they were concerned, and 32% said they were not.  (Another 17% were in the middle, plus the 9% who did not respond to the question.)  Not surprisingly, more Republicans (61%) said they were concerned than Democrats (31%).

In contrast, when I asked the identical question in the CCES right before the 2016 election, 37% were concerned and 40%, were not. This shift in opinion across the year — 6 percentage points more saying they were concerned and 8 points fewer saying they were not — is small, but statistically significant.

One interesting detail in this shift is that Democrats, more than Republicans, have moved in the direction of being concerned about fraud over the past year.  A bit over a year ago, 21% of Democrats said they were concerned about vote fraud, compared to 31% last week.  In contrast, 57% of Republicans said they were concerned about voter fraud in 2016, barely any different from the most recent results.

It’s not obvious why Democrats would state they were more worried about fraud across the past year and not Republicans.  One possibility is that Democrats and Republicans have different ideas about what constitutes voter fraud when considered in the abstract.  Democrats may be more likely to think about Russian hacking while Republicans may be thinking about double-voting and non-citizen voting.  Still, there’s a built-in asymmetry in the debate about voter fraud, however it’s defined.  It’s easier to illustrate cases of vote fraud, even when they’re rare, than to illustrate the lack of fraud, similar to the problem of proving a negative.  Plus, Republicans were already much more concerned about voter fraud than Democrats, so that Democrats could more easily be moved in the “worried” direction.

In any event, it looks like the public has moved slightly in the direction of being more concerned about fraud.  Whether that movement was caused by the commission’s work is unclear.  (I suspect the direct causal effect was slight.)  Had the commission been able to continue its business, and had been able to maintain a unified narrative (which is a big “if”), it does appear that public opinion was trending in the direction that was sympathetic to the commission leadership’s policy agenda.

 

Tomorrow’s post: Knowledge about the Commission

Small blue shift in Alabama canvass

A few years ago Ned Foley identified a growing trend in post-election canvasses:  there is a tendency, at least in presidential elections, for the Democratic share of the vote to grow between the initial election-night count and the final, official canvass of the vote.  He referred to this as a “big blue shift,” paying homage to the association of the Democratic Party with the color blue.  Ned and I have written together on this, including a conference paper that I hope we can resurrect soon, and a Monkey Cage/Washington Post op-ed we published on the morning of the 2016 election.

While there is no claim that the blue shift will show up in every election, changes to election laws over the past two decades have created a set of ballots that require special handling, and which are often only resolved in the days and weeks following the election.  The largest set of these ballots are absentees, which have been weakly (and inconsistently) trending Democratic over time, followed by provisional ballots, which are much more regularly Democratic.

Therefore, it’s not surprising to see that Doug Jones benefited by a very small blue shift in the Alabama vote count following the recent special election.  The election night count had Jones at 49.92% of the vote, to Moore’s 48.38%, or a difference of 1.54 points.  The official canvass, which has just been certified, had Jones at 49.97% of the vote, to Moore’s 48.34%, or a difference of 1.62 points.

This is a tiny difference, but it was pervasive in the state, as the accompanying graph shows.  (If you want to read the details, I’m afraid you’ll need to print it out and use a magnifying glass.  Click on the graph to enlargify it.)

Here’s how to read the graph.  Alabama’s counties are sorted according to their election night percentage for Jones, with the most pro-Jones counties on top.  Deviations from the election night percentage are shown.  The dashed line directly to the right of the 0% line is the statewide average shift toward Jones.

It’s easy to see here how relatively uniform the shift was across the counties.  The Jones percentage of the vote increased in 54 counties, declined in 7 and didn’t change in the remaining 6.  In only 8 counties was the magnitude of the difference greater than a tenth of a percentage point, and only one of those (Mobile) could be considered a large county.  (Jefferson County, the state’s largest county, which got beat up in Roy Moore’s filing attempting to delay certifying the vote, came out almost exactly at the state average.)

Why the blue shift occurred in this particular election is still an open question.  I haven’t seen a full accounting of domestic absentee, UOCAVA, and provisional ballots yet, but would suspect that provisional ballots dominated in this case.  In Mobile County, at least, absentee ballots and provisional ballots came in 70.2% and 81.8%, respectively, for Jones, compared to Jones’s overall county vote of 57.3%.  But, that’s just one county.

Election officials often say that their job isn’t to get the vote count fast, but to get it right.  Here we see in Alabama a small example of how vote totals shift from election night to the final canvass.  Given the controversy surrounding the election, it was probably a blessing that Jones had won the election night count, even by a little, so that the blue shift didn’t enter into the post-election count.  One can only imagine what would have happened had Moore been ahead on election night…

Much ado about nothing in Alabama “fraud” charges

At the risk of being lost down a rabbit hole and subject to endless trolling, I just have to weigh in on the so-called evidence of vote fraud that was contained in Roy Moore’s court filing, in which he tried to get a delay in having the vote certified.  (The reason I decided to plow ahead is that Moore’s filing points out an interesting pattern in the precinct returns — it’s just that it’s not evidence of vote fraud.)

There are a lot of claims made in Moore’s filing, and I don’t pretend to have time to take them all on.  The one that has the look of seriousness is based on some number crunching by Philip Evans, an electrical engineer from South Carolina who has taken a look at the precinct-level election returns from Jefferson County (Birmingham) and declared them to be impossibly skewed — or, as Mr. Evans  puts it, based on analyzing more than one hundred elections, “never has there been the level of statistical proof on the scale of Jefferson County” that the results were fabricated.

What’s this proof?  It’s nothing more than calculating the percentage of voters in a precinct who voted for Moore and then calculating the percentage of voters in a precinct who voted “Republican” for the straight ticket line.  (Recall, that it was possible to vote for Moore in Alabama “without voting for Moore,” by simply voting Republican on the straight-ticket line.  Such a vote would mean a vote for every Republican on the ballot — which in this case was simply Roy Moore.)

The analysis then simply shows the distribution of this difference (Moore % – Straight Republican %) for each precinct.  Here is a screen shot of the graph Mr. Evans uses. (Click on the graph to enbiggen it.)

Every precinct on the left side of the graph had a higher percentage of voters vote straight Republican (among those who voted the straight party line for any party) than voted for Moore (among those who voted directly for the candidates themselves.)

For starters, the gist of the argument is that the figure above doesn’t follow a normal distribution, which alone is evidence of fabricated election results.  I know of no reason the distribution should follow the normal distribution.  Most of the political world isn’t distributed normally (or Poisson, or uniform, or any other favored distribution).  That’s a red herring.

The real issue is whether the relationships between variables look anomalous.  The graph above didn’t answer the question about relationships, so the first thing I did was to recast it in a way that explores the implied relationship among variables that are conflated when we just gaze at a single distribution.  So, I created a graph of the % of the vote that Moore received purely as a candidate against the % of the vote that the Republican Party got on the straight-party line.  This is what that graph looks like. (Click to enlarge.)

The data are as described in the previous paragraph.  The solid gray line is the 45-degree line.  The dashed line is a non-linear best fit to the data.  (For those keeping score, this is a lowess fit.)

Note that the red precincts — the one Moore called “anomalous” — are relatively competitive.  In fact, given that Jefferson is a Democratic county, these are relatively strong Republican precincts for the county.

So, it appeared that Moore as a stand-alone candidate under-performed in moderately Republican precincts in Jefferson County, compared to those who voted a straight ticket.

The Moore court filing suggests that patterns like this can’t be explained.  Quite the contrary.  The precincts that the Moore campaign seem so worried about are precisely the types of places that commentators were speculating about ahead of the election:  solidly Republican areas that just couldn’t stomach voting for Moore.  Also, because these are precincts where party loyalties are closely balanced, I would bet that they are full of voters that political scientists call “pivotal,” that is, voters who could go either way in an election.  Again, these were precisely the voters who were up for grabs in the special election.  And, it’s also these precincts where you might expect to find more Republicans “voting the party, not the man.”

Finally, there’s the empirical issue of how correlated “candidate-only votes” and “party votes” are at the precinct level in a less controversial election.  Turns out, they’re not as highly correlated as you would expect.  For instance, the following analysis reproduces my scatterplot, this time using the precinct election returns from Jefferson County in the 2016 presidential election.

Here, I’ve highlighted “outlier” precincts in red, using the same criterion as before.  (I.e., the red precincts are more than 20 points away from the diagonal line.)  Unlike the special election, there is a bit more symmetry on both sides of the line.  Still, Trump as a candidate was more likely to under-perform the Republican straight-ticket line than Clinton was to under-perform the Democratic line.  And, the under-performance was more common in precincts that moderately leaned one way or the other — again, the type of precinct where you might find more wavering partisans, because of the context in which they live.

So, the pattern that was used by the Moore campaign to claim fraud was different in details from 2016 in Jefferson County, but only in the details.  In 2016, both parties arguably had flawed candidates, and thus it’s not surprising we would see the parties’ candidates under-perform their party in a few precincts.  Nor is it surprising that the precincts where they under-performed would be relatively central to the overall partisan space.

To conclude, I’m the last person in the world to say you shouldn’t mine through election data looking for evidence of erroneous election returns, either due to fraud or human error.  But, finding what appears to be anomalies is just the beginning.  The next step is that the person suspecting fraud owes it to everyone to consider alternative explanations.  In the case of the 2017 Alabama special election, what appears to be anomalous is easily explained in terms of well-known campaign dynamics — and may not be as anomalous as it first appears.  Finally, any charge of fraud (or human error) needs a mechanism to hang the story on.  There was just no plausible mechanism proposed in Alabama.  (The implied mechanisms were so out-there that they don’t deserve to be mentioned.)

Doug Jones won in a fair fight.  If the Republicans want to win the seat back, they’ll have to get it back in a fair fight.

 

Interesting residual vote pattern in Alabama

Many people know that I’m interested in the residual vote rate, which is the  percentage of ballots that do not have a countable vote for a particular race.  A residual vote is either an over-vote or an under-vote.  Residual votes were at the core of the 2000 recount fiasco in Florida, where hanging or pregnant chad led to under-votes, and where weird ballot designs such as the “caterpillar ballot” caused over-votes.

What did the residual vote rate look like in the Jones/Moore race in Alabama?  According to preliminary election results published by the Alabama Secretary of State’s office,  of the 1,346,146 Alabamians who went to the polls last Tuesday, 1,779 cast no vote in the Senate contest.  That’s 0.13% of turnout, which is a pretty small number.  Still it’s not zero.

Why would a voter go to the polls in a hotly contested race and not vote in the contest?  There are at least four possibilities:  (1) The voter felt a civic duty to turn out, but couldn’t bring himself to vote for one of the candidates.  (2) The voter made a mistake when marking the ballot and didn’t leave a legal mark.  (3)  The machine failed to read the legal mark.  (4)  The voter turned out to vote in another contest, and abstained in the hot contest.

It turns out that reason # 4 was a major driver of the residual vote in Alabama last Tuesday.  We can see this in the following graph, which plots the residual vote rate for each county in the U.S. Senate special election.  (Click on the graph for a bigger view.)

I have labeled the nine counties that are outliers.  What sets them apart from the rest of the counties?  With the exception of Choctaw, they all had other things on the ballot, generally taxation or millage questions.  Considering how salient issues of local taxation can be, it’s not surprising that some Alabamians in the eight counties with referendum items would have shown up to vote about taxes, but not about the Senate.

How many voters might this be?  No many.  A back-of-the-envelope calculation suggests that only about 127 voters showed up in these 8 counties to vote for taxes but not for senator.

Similarly, it’s possible to use simple statistical techniques to estimate approximately what fraction of the residual vote rate was due to these additional ballot questions, or similarly, what would the residual vote rate have been if no counties had put tax questions on their ballots.  The residual vote rate In the 59 counties without tax questions was 0.06%, compared to 0.51% in the eight other counties.  A little algebra suggests that about half of the statewide residual votes in the Senate election were due to voters showing up for tax questions and skipping the Senate rate.

One final word about voting machines:  The residual vote rate, if used properly, can be a valuable tool to diagnose inaccuracies in voting machines.  One way this can be done is just by comparing the residual vote rate across different machines.  In Alabama, if we do that with the Senate race, we see that counties using the DS 200 scanner had a significantly higher residual vote rate than counties that used the Model 100s (0.17% vs. 0.07%).  However, this is entirely due to the fact that most (7/8) of the counties that had tax questions on the ballot used DS 200s.  Controlling for whether a referendum was on the ballot in a county reduces the difference between scanners to virtually zero.

None of this, by the way, explains what happened in Choctaw County, which was the only outlier county with no tax questions on the ballot.  It’s a very small county (4,240 voted), so that a small number of residual votes (18) can produce a residual vote rate that stands out by comparison.  This might just be a “law of small numbers” issue, but it also might reflect something the county probate judge might look into.

Choctaw County, the only outlier with no tax questions on the ballot, is explained by the fact that the election results posted on the Secretary of State’s website don’t report any write-ins.  However, the results reported by AL.com do show 17 write-in votes in Choctaw.  So, it’s not an outlier after all.  (Thanks to Justin Levitt for pointing out my error here.)  Lowndes County also doesn’t have any write-ins recorded on the SOS’s website, although AL.com does.  However, Lowndes also had three school tax questions on the ballot.  Accounting for the 13 additional write-ins in Lowndes leaves it an outlier, just less of an outlier.