Get Out the Vote: How behavioral science can boost voter turnout

In this special episode, we discuss how behavioral science can be used to encourage voter registration and turnaround. We cover well known ideas like social proof including some of the most relevant nuances and how it’s often misused by campaigns for good. We also turn to lesser known research by Bryan on how we label behavior can have a significant effect on whether it occurs.

Transcript

MAF: [00:00:00] Welcome back to Behavioral Science for Brands, a podcast
where we connect academic insights and practical marketing to help grow your
business and your brand. Every other week, Richard and I sit down and talk
about some of the country's best brands and the behavioral science that powers
them. I'm MichaelAaron Flicker.
RS: And I'm Richard Shotton.
MAF: And today we're talking about the very important idea of getting out the
vote. As it's election season here in the United States, we're going to look at
ways behavioral science can help strengthen civic participation. Let's get into it.
So Richard, as we get into today's episode, we thought we would do something
a little bit more timely for Americans as they're getting ready for the November
5th presidential elections.
This episode will go live in October and we're thinking about what are the
things that cause people to vote and what are ways [00:01:00] behavioral
science can help get more civic participation and to set the groundwork for
where we find ourselves in America - America has about 66 percent of the
population that votes. That may be a surprise to many of our listeners who
might think that there's a much higher percent that votes.
In fact, the United States is 31st amongst all developed nations. And not as high
as maybe we'd like to see. On the other hand, the 2018, 2020, and 2022 all set
records for highest voter participation. So, on the one hand, not as high as some
of the most - as some of the countries that have the most voting participation,
like New Zealand, Germany, and Canada who all have over 90 percent of their
democracies vote.
In the 2020 [00:02:00] election, 62.8 percent was the highest in a presidential
election since 1900. So there's a bunch of things going on here, that we will
eventually dissect, but I thought it'd be interesting to talk a little bit about some
historical get out the vote campaigns that have occurred. I picked one in the
United Kingdom and one in the United States to give everybody a sense of what
types of efforts have gone on.
And then we can really dive into some behavioral sciences powering them. So
one of the most famous Voter registration drives that's occurred in America is
something called Rock the Vote. It started in the 1990s, and it's still a program

and a campaign that's run today. And the original idea was to engage young
voters and get them more excited to come out and vote.
So, in the nineties, there was a very big collaboration with MTV to reach young
voters and really get them active, [00:03:00] using music and celebrities.
Getting them to find voting more appealing. It was credited in the 1992
presidential elections with increasing youth turnout by 20%. So aligning with
culture, aligning with celebrities, drove the Rock the Votes early wins to get,
uh, youth participation up.
In the UK, in 2005, there was a campaign very similar called Got Five. And the
Got Five campaign was was all about emphasizing that getting a registered to
vote just takes five minutes. And so, can we spare five minutes to get you to be
more civically involved? That campaign in 2005 drove a 6 percent increase in
voter registration.
So we have these examples of developed nations like the United States and the
United Kingdom wanting to get increased voter participation. The first step in
America [00:04:00] is voter registration, which then leads to actually voting.
It’s an interesting history going back a hundred years to why the United States
requires registration separate from, let's say, getting your license. Other
countries, you get your license and your voting registration at the same time.
The United States, it's a separate act. But regardless of why that is, a lot of the
efforts in the United States are centered around getting you to register.
And if you register, then you're much more likely to vote. So that's the focus for
today. As we think about behavioral science and how it can be applied to
increase voter turnout and strengthen democratic participation, you and I
thought we might focus on two separate areas. The first is using positive social
proof to help frame folks getting more involved in voter registration.
And the second is the principle of consistency and how small language tweaks,
like using nouns instead of verbs, can help increase [00:05:00] the likelihood of
voting. So let's get into it. Let's start with this idea. We're not just talking about
social proof, which we've covered in other episodes. We're talking about
focusing on positive social proof.
Maybe you could talk a little bit about that.
RS: Yes. There's, there's two ways of using social proof. The ideal way is to
emphasize the scale of the behavior you want to encourage. So if you were

going to use social proof in America, you could say, 200 million people vote
each year or 150 million people vote each year.
That would be emphasizing lots of people vote.
MAF: Shouldn't you?
RS: Yeah. It will make the behavior more appealing. People are more likely to
do it. Because if so many people are voting you, you think to yourself, well, if
everyone else is doing it, you have to at least consider whether it's the right
thing for you.
Unfortunately, when it comes to behaviors like registering to vote, in fact, an
awful lot of social good campaigns, well-meaning communicators inadvertently
make the situation worse by often [00:06:00] emphasizing the scale of the thing
they don't want. Emphasizing the scale of non-voting. So if you said only 60
percent or 50%, 50, 60, 70 million people don't vote every year.
That would be what I would call negative social proof. You are emphasizing the
scale of the unwanted behavior and that's a mistake. It tends to backfire.
MAF: And even if you don't do it literally, when you used your example, 150
million people vote. Well, I know there's over 300 million people in America.
So right away I say, Oh, sounds like there's a lot of people who don't vote.
RS: Well, I mean, I don't know on that one. I think the evidence would suggest
the immediate reaction is 150 millions, a large number, but you, I think you're
right in that, you could create a negative social proof problem even by
emphasizing, any lack of registration. Now that that's not speculation.
There's a really good 2003 study from Robert Shieldini. So he's at Arizona State
[00:07:00] University. And he wanted to know, what happens if you emphasize
the scale of theft that was happening in a national park? So he works with the
Arizona Petrified Wood National Park, where loads of people are stealing bits
of petrified wood as a souvenir.
He goes to this park, finds a deserted path and puts up hidden CCTV cameras
pointing at the path. And he sprinkles bits of petrified wood along the path. So
anyone who walks down this. Can't miss it.
MAF: Can't miss it.

RS: Exactly. They've got the opportunity to steal. Now, when there's no sign up,
2. 9 percent of people steal a bit of wood.
Um, that's his baseline. When he puts up a sign saying, don't steal, it's wrong.
Theft rates dropped to 1. 7%. So he's shown you can intervene in this behavior.
This is something you can change for the better.
MAF: So far so good.
RS: But then the third and final intervention, he puts up [00:08:00] signs saying
14 tons of wood have been stolen each year, and it's ruining the look of the park.
Don't steal, it's wrong. Now, when he does that, there is a almost tripling of theft
rates. We're up to 7. 9 percent of people stealing a bit of wood. So in Cialdini's
words, this is a crime promotion strategy, not a crime prevention strategy. What
he's shown here is if you emphasize the scale of the problem, you remove a
sense of transgression and people become more likely to indulge in the behavior
that you don't want.
This translates to voting. Uh, back in the 2015 election, the government ran a
campaign aimed at encouraging female participation.
MAF: This is in the United Kingdom.
RS: Yes, in the United Kingdom, absolutely. And the slogan was there's
missing millions because the female participation rate in voting was slightly
lower than the male.
Now that emphasis on millions of people not voting would tap into [00:09:00]
negative social proof. It would make it more likely women didn't vote. What
they could have focused on, equally honestly, was the fact that for the last three
elections, female voting rates had been going up. This is never an excuse to
make up statistics and lie, but you can often cut exactly the same data to
emphasize a different situation.
So, don't talk about the millions not voting talk about the millions who do vote.
Focus on positive social proof, not negative social proof.
MAF: Because we want to insert in the mind of the recipient of the message
that this is going in the right direction?

RS: Exactly. Either the behavior you want to encourage is commonplace and
regular, or you could even just emphasize that it's growing.
Right. So that's known as dynamic social proof. There’s a 2017 study from
[00:10:00] Sparkman and Walton at Stanford that demonstrates it very nicely.
And they worked with a cafe in the U S and over a number of weeks, they
alternated the signs that were on the dinner tables, and sometimes they said, ,
basically, three out of 10 Americans are trying to limit their meat consumption.
So they emphasized meat-free eating was a minority behavior. And when they
did that, 17 percent of the dishes were meat free that were ordered. Other
occasion, same cafe, basically the same message, but it was now prefaced with
more and more Americans are trying to limit their meat consumption. In the last
few years, three out of 10 have started to eat less meat.
MAF: So they use “started” versus the opposite.
RS: More and more and started. They're emphasizing ‘yes, it's still a small
behavior, but it's growing. There's momentum.’ And there was an exact
doubling of meat-free ordering. So you now have 34 percent of people ordering
a dish that didn't have meat in. [00:11:00] They call this dynamic social proof.
And essentially it's the argument that you don't even have to have absolute scale
to harness social proof. If you can honestly emphasize that the behavior that you
want to encourage is growing in popularity, there's momentum, there's
dynamism, you'll tap into social proof very powerfully. So you mentioned right
beginning of the episode, the statistics on increasing vote participation.
Uh, emphasizing that positive change from the prior election would tap into
dynamic social proof.
MAF: And our conversation today is around politics, but you see this being
used in e-commerce stores. This product's being bought a lot, frequently, 39
people have purchased in the last day. You see that even if it's a small number,
the growing trend, the dynamic social group being used in e-commerce pretty
regularly.
RS: Absolutely. It is a common tactic in e-commerce. And I always find it
interesting that certain [00:12:00] areas or sectors might use biases very
regularly. And social proof is regularly used in e-commerce. But other areas
within marketing shun exactly the same bias. So if you look at campaigns for

good, charity campaigns, government campaigns, ethical campaigns, they often
emphasize the scale of the problem, the scale of the unwanted behavior.
So in this particular situation, I think get out of the vote campaigns could learn.
From e-commerce campaigns.
MAF: Yeah. And what it teaches all of us who have brands or businesses is to
think about, when we frame a certain challenge or certain opportunity, how do
we frame it in a way that best helps us get to our outcome that frames it
positively or frames it as growing so that you can best get to the outcomes that
you want.
RS: Absolutely.
MAF: So let's go to a break. And when we come back, let's talk about the
difference between noun and verbs. And if our listeners at home, we have to
remind the definitions, we will. And [00:13:00] then, uh, we can get into that
part of the discussion.
Auto: Behavioral Science for Brands is brought to you by Function Growth,
AdAge’s 2023 Newcomer Agency of the Year.
Your DTC business isn't broken. You're just doing it wrong. By fusing
advanced technology and behavioral science, Function Growth solves the
biggest problems faced by today's DTC brands. Visit us at
www.FunctionGrowth.co to learn more.
MAF: Welcome back to the Behavioral Science for Brands podcast, where we
connect academic insights and practical marketing to help grow your brand and
your business.
And today we're talking about get out the vote. And we're specifically thinking
about the November presidential elections here in the United States and how
what we use commercially to help grow brands and grow their businesses also
has a lot to teach us about how we engage and increase civic participation in
democracies.
So before the break, Richard, we talked about positive social proof and dynamic
social proof. [00:14:00] Now we want to shift gears and talk about the
difference between a noun and a verb. The word choice you make changes the
way that you think about the topic.

RS: Yeah. So there's a lovely study from 2011 by Christopher Brian, who's at
Stanford, and he recruited a group of people, this is in the run up to the 2008
presidential election, recruited a group of Californians and sent them a survey.
And in some of them he said, how important is it to you to vote at the upcoming
election? Other people, he said, how important is it to you to be a voter at the
upcoming election?
MAF: So the difference is…
RS: First group, verb, second group, noun. And what Brian finds when he looks
at state records after the election, the people who heard the noun, the voter word
were more likely to take part in the election than the people who heard the verb.
So you hear the noun, you're more [00:15:00] likely to vote than hearing the
verb. Brian's argument is that there is a fundamental difference between the
ways of phrasing the question. If you use a verb, that's something people do. If
you use a noun, that's something that people are. It might feel a bit of a big leap
with this discussion of voting, but think about it on a more personal level.
If I were to say MichaelAaron is a tennis player, it would feel that tennis is a
massively important thing to you. If I said Michael Aaron plays tennis, it just
feels like it's something you
MAF: It's a hobby.
RS: It's a hobby that you happen to do occasionally. We believe that this noun
form reflects the essential self, the verb just reflects a kind of transient action.
That has been used successfully by Brian in voting situations, but I think the
really interesting bit here [00:16:00] is as a brand you can apply this in
commercial situations. Imagine you're a big magazine company, and you get
people to subscribe to your magazine. When that annual subscription comes to
an end, just beforehand most magazines will send out a letter saying, ‘thank you
for subscribing, please subscribe again.’
The simple behavioral science tweak that would cost you nothing extra is to
change that opening sentence to say, ‘thank you for being a subscriber, please
subscribe again’ because people want to be consistent with their past behavior.
And if they're subscribers, that feels like an important action that they
undertake.

It's important to their identity. If you just describe it as a verb, it's much less
meaningful. And that bias of consistency is less tapped into. So, Yes, we can see
the studies done by academics in voting situations, but it doesn't take much of a
leap to think about [00:17:00] how you can apply it in a commercial setting.
MAF: It strikes me that the idea that this is about identity and not action is
really at the center of the insight. What else can we learn about this idea of
describing identity, talking about identity? As it relates to voting or to brands
and businesses, like what's your reflection on this uncovering and naming
identity things?
RS: I think my main reflection is that many of these behavioral science
experiments are costless to implement. Now, if you're the magazine trying to
encourage renewals, you've got to send out letters. You have to use a certain
selection of words. Why not, make sure that you incorporate the words that
have been proven to be most effective.
It doesn't cost you any extra to use a noun over a verb, but the evidence from
Brian and others suggests it will increase the effectiveness of your
communications. So I love the fact [00:18:00] that these interventions can often
have a very, very low cost or they're completely costless.
MAF: Yeah. And I think sometimes we hear behavioral science techniques
because they're small.
Because they sometimes are called nudges or people think of it, they're
relatively inconsequential, but this is where the idea of stacking - This is where
the idea of every small gain can lead to better and better outcomes. There’s only
a cost to not do it. But there's no incremental cost to do it.
RS: Yes, and some - I think we've got to be careful though, that we sometimes
think that a large effect requires a large intervention. And one of the findings
that comes up again and again in behavioral science is seemingly small tweaks
can often have
MAF: Big outcomes
RS: Big changes. So I think you're absolutely right to categorize noun versus
verb as a minor costless change that you should put in place along with lots of
other things to get a combined [00:19:00] cumulative big effect.

But if we go back to the very first study we talked about, the negative social
proof study, talking about “Don't steal it's wrong” 1. 7 percent of people stole a
bit of wood saying, “14 tons of wood have been stolen each year and it's ruining
the look of the park. Don't steal.” It was 7. 9%. I mean, you've got this massive
swing in behavior from a simple and seemingly very minor intervention.
I think sometimes the biases don't get used enough because people think, “Oh,
that tiny little change can't possibly affect my really big serious business
problem.” I would just say the evidence suggests otherwise. Have the
confidence to test these ideas on your brand and you'll see the impact they can
have.
MAF: So helpful. So if we start to think about noun versus verb, if we focus on
that side of the discussion, is there any concerns about trying to name the
identity of the person you're talking to? When you say, “I [00:20:00] don't read
the economist, I'm an economist reader”, it feels relatively low risk. But are
there other instances where trying to use a noun could lead to more risk?
I'm thinking about if I'm a brand, people may receive this and say, “Oh, I'm not,
you know, I don't know if I agree with that”. Any concern on that front?
RS: I think that's a fair challenge. I think, with all these Biases when it comes to
the practical marketing application, there's science and art. The science is
people will identify more with a noun than a verb.
The art is when is that the most appropriate place to introduce the intervention?
And I think you're right. If people feel that the communicator is labelled in a
way that is inauthentic and doesn't represent them, there's more danger from a
noun than a verb because it will feel crass. So I think you're right to say that
context [00:21:00] is important and be careful where you're applying these
ideas.
But I think that's one, an area of judgment. And then secondly, it probably
emphasizes the importance of when you first introduce a behavioral science bias
into your communications, test it on a small scale, see what the implication is, if
it doesn't work, you turn it off, minimal downside. If it does work, you run it
forever on a mass scale and there's a massive upside.
So, the odds are stacked in your favor with these interventions.
MAF: Yeah. And I would say, uh, in addition to testing, you're not going to get
the maximum outcomes, uh, from your creative and from your activities, if you

don't continually try to add and innovate those things. So this is just another
way to think about how to change your communications to have more
resonance.
So Richard, as we come to a close, could you remind us the big topics we
discussed today?
RS: So I've discussed three big things. The first is the idea of [00:22:00] social
proof. It's one of the most robust findings in behavioral science. It's the idea that
we are a herd species and we're deeply influenced by what others do.
So if you want to encourage your behavior, emphasize how many people are
behaving in that way, but also, especially in the area of voting and social good
campaigns, be very careful not to use social proof in the wrong direction, what
we can call negative social proof. Do not emphasize the scale of popularity of
the unwanted behavior.
If you stress lots of people don't vote, you'll make the situation worse. So focus
on the popularity of the desirable behavior, not the scale of the unwanted
behavior. That's the first point. The second point was you can harness social
proof even with behaviors that are happening on a reasonably small scale.
If you draw attention to the change in volume of behavior. So this is known as
dynamic social proof. We talked about the cafe study with Spartan and Walton,
and it's the idea if you emphasize a [00:23:00] behavior is growing in
popularity, even if the absolute volume is small, emphasizing momentum and
dynamism is a very effective way of tapping into social proof.
And then the third and final study, this is the idea that if you describe people in
a noun form. MichaelAaron is a tennis player. It's more associated with their
identity than if you use a verb. MichaelAaron is - sorry, MichaelAaron plays
tennis. You can use that to your benefit. If you want people to continue with a
behavior, describe them as a subscriber in the magazine example, not as having
subscribed.
So if you want to get a behavior continued, prioritize nouns over verbs.
MAF: Perfect. So as we come to a close, Richard, we've been talking about
voting to encourage everyone to think positively about voting. Can you
remember the first time you voted? Where were you? What was happening?
You don't have to share who you voted [00:24:00] for, but what was the
situation?

RS: Uh, so 1997, I would have been 21, uh, I was at university and we had had
the conservative party in power. Wow, since 1979, they've been in power for 18
years. And, uh, 1997, Labor won by an absolute landslide. And I can remember
that sense of, um, you know, as a young person, you're often a vote more left
wing, that sense of euphoria that there was this, uh, new change coming in.
And I think many times in politics, nothing quite lives up to expectation would
be a fair way of putting it.
MAF: That's funny. That's funny.
RS: And then for you, what was the, what was the first election you voted?
MAF: I was voting for the first time in 2000, which was a presidential election.
My birthday is October 23rd and the election is almost a week later.
So I had to go in front of a judge to get approved to [00:25:00] vote because the
time was so short. So you go into the courtroom, you say, it's my, you know, it's
time to - you know, I'm 18 years old. I'm eligible to vote. And anyway, that, that
was what got me in to be able to vote. In those November elections, but as you
shared the hope and the euphoria, uh, 2000 election in America was very, uh,
contentious because it was the
RS: That was the George W. Bush hanging chads. Correct. Wow.
MAF: Correct. It was down to hundreds of votes that turned the election. Um,
but I was glad I voted. And I always remember, uh, trying to explain, you know,
it was not a, it was not a big moment to the judge, but it was a big moment to
me to stand up and ask to be able to vote. So lots of positive memories with
that.
RS: Fantastic.
MAF: And good for everybody to remember, democracies only work when you
have participation. We need people to get out the vote, to encourage young
people to vote and for everybody to have a voice. Well, that wraps up today's
episode. If you found our discussion [00:26:00] valuable, we'd ask that you
follow us on YouTube or leave a review and comment.
That helps us reach more listeners just like you. And if you'd like to dive deeper
into the studies we talked about today or learn more, visit
www.theconsumerbehaviorlab.com. There, we have a video of today's episode,

a full transcript and the show notes, which give you links to the resources and
the original source for materials that we discussed.
We also are on LinkedIn. Please follow us on LinkedIn to stay up to date on all
of the newest news from the Consumer Behavior Lab. And until next time, I'm
MichaelAaron Flicker.
RS: And I'm Richard Shotton.
MAF: Thanks for tuning in and we look forward to bringing you our next
episode of Behavioral Science for Brands week after next.