WEBVTT - Do Political TV Ads Really Change Voters' Minds?

0:00:01.800 --> 0:00:07.520
<v Speaker 1>Welcome to Brainstuff, a production of iHeart Radio. Hey brain Stuff,

0:00:07.600 --> 0:00:12.840
<v Speaker 1>Lauren Vogelbaum here. The two mid term elections are fast

0:00:12.840 --> 0:00:15.720
<v Speaker 1>approaching in the United States, and if you live in

0:00:15.760 --> 0:00:18.880
<v Speaker 1>a state with a hotly contested race and watch television,

0:00:19.120 --> 0:00:24.759
<v Speaker 1>you're probably being blasted by political TV ads. By late September,

0:00:25.079 --> 0:00:28.080
<v Speaker 1>campaigns across the United States had already spent more than

0:00:28.120 --> 0:00:32.080
<v Speaker 1>six point four billion dollars on ads, including television, print,

0:00:32.120 --> 0:00:35.280
<v Speaker 1>and online, and are expected to spend a total of

0:00:35.400 --> 0:00:38.800
<v Speaker 1>nine point seven billion by election day in November, far

0:00:38.880 --> 0:00:44.240
<v Speaker 1>more than either the eight or twenty elections. This year's

0:00:44.240 --> 0:00:47.400
<v Speaker 1>record spending on TV and other types of political ads

0:00:47.560 --> 0:00:51.360
<v Speaker 1>reflects the high stakes of the two mid terms. The

0:00:51.479 --> 0:00:54.200
<v Speaker 1>Senate is split fifty fifty and the Dempcrats hold the

0:00:54.240 --> 0:00:58.680
<v Speaker 1>House of Representatives by a slim nine seat majority. For

0:00:58.800 --> 0:01:01.120
<v Speaker 1>the article this episode is based on, has to Fork

0:01:01.200 --> 0:01:04.320
<v Speaker 1>spoke with Christopher Warshaw, a political science professor at George

0:01:04.360 --> 0:01:08.880
<v Speaker 1>Washington University. He said campaigns hope that by buying a

0:01:08.880 --> 0:01:11.560
<v Speaker 1>few extra thousand ads, they can put their candidates over

0:01:11.600 --> 0:01:14.600
<v Speaker 1>the top and win control of the government for their party.

0:01:14.840 --> 0:01:18.000
<v Speaker 1>The American public is more polarized than ever before. But

0:01:18.080 --> 0:01:20.920
<v Speaker 1>there's still the swath of independent voters in the middle

0:01:21.000 --> 0:01:25.479
<v Speaker 1>who are trying to decide, and political ads, despite how

0:01:25.520 --> 0:01:29.200
<v Speaker 1>annoying and even stressful they can be, have proven effective,

0:01:29.440 --> 0:01:33.440
<v Speaker 1>especially in the tightest races. Even a bombardment of ads

0:01:33.560 --> 0:01:37.800
<v Speaker 1>doesn't necessarily turn people off. Warshop and two other researchers

0:01:37.840 --> 0:01:40.480
<v Speaker 1>published a study this year in the journal American Political

0:01:40.520 --> 0:01:45.119
<v Speaker 1>Science Review examining the effectiveness of political tv ads. They

0:01:45.160 --> 0:01:49.080
<v Speaker 1>found that basically the one thousand that is only modestly

0:01:49.160 --> 0:01:55.200
<v Speaker 1>less effective than the tenth presidential election was decided by

0:01:55.280 --> 0:01:58.720
<v Speaker 1>razor thin margins of victory for Republican candidate Donald Trump

0:01:58.840 --> 0:02:04.560
<v Speaker 1>in battleground states like Michigan, Pennsylvania, and Wisconsin. In Joe

0:02:04.600 --> 0:02:07.680
<v Speaker 1>Biden and the Democrats spent heavily on political tv ads

0:02:07.720 --> 0:02:13.400
<v Speaker 1>in those same battleground markets. As election day of approached,

0:02:13.600 --> 0:02:16.959
<v Speaker 1>pro Biden tv ads outnumbered pro Trump ads by over

0:02:17.040 --> 0:02:21.200
<v Speaker 1>five thousand in key markets like Harrisburg, Pennsylvania, and by

0:02:21.240 --> 0:02:25.880
<v Speaker 1>over four thousand in Green Bay, Wisconsin. In their research,

0:02:26.000 --> 0:02:29.320
<v Speaker 1>Warshaw and his colleagues estimated that for every additional thousand

0:02:29.360 --> 0:02:32.480
<v Speaker 1>TV ads a campaign runs compared to its opponent, it

0:02:32.560 --> 0:02:36.200
<v Speaker 1>can capture zero point two percentage points more of the vote,

0:02:36.840 --> 0:02:40.000
<v Speaker 1>and the results of the presidential election seemed to bear

0:02:40.040 --> 0:02:45.080
<v Speaker 1>this out. In sixteen, Trump one Wisconsin by about zero

0:02:45.200 --> 0:02:49.600
<v Speaker 1>point seven percent, but in Biden one Wisconsin by zero

0:02:49.680 --> 0:02:53.600
<v Speaker 1>point seven percent. The same swing happened in Pennsylvania, in

0:02:54.560 --> 0:02:57.919
<v Speaker 1>Trump won by zero point seven percent, but in Biden

0:02:57.960 --> 0:03:02.240
<v Speaker 1>won the state by one point two percent. For their study,

0:03:02.320 --> 0:03:05.079
<v Speaker 1>Warshop and his co authors analyzed data for more than

0:03:05.120 --> 0:03:07.840
<v Speaker 1>two thousand, two hundred and fifty U S elections from

0:03:07.880 --> 0:03:12.000
<v Speaker 1>the year two thousand through and not just presidential races.

0:03:13.040 --> 0:03:15.320
<v Speaker 1>What they found was the tv ads had a much

0:03:15.360 --> 0:03:19.480
<v Speaker 1>greater effect on down ballot races compared with presidential contests,

0:03:19.919 --> 0:03:22.960
<v Speaker 1>three times greater effect for senate races, four times greater

0:03:23.000 --> 0:03:26.200
<v Speaker 1>for gubernatorial races, and twelve times greater for the more

0:03:26.240 --> 0:03:31.160
<v Speaker 1>obscure races like state treasurer. The reason, says Warshop, is

0:03:31.200 --> 0:03:33.520
<v Speaker 1>that the whole function of political tv ads is to

0:03:33.560 --> 0:03:37.520
<v Speaker 1>convey some sort of new information to the voter. Quote

0:03:37.920 --> 0:03:40.320
<v Speaker 1>at the presidential level, ads don't give you much new

0:03:40.320 --> 0:03:43.520
<v Speaker 1>information that you didn't already have. We already knew so

0:03:43.640 --> 0:03:46.360
<v Speaker 1>much about Donald Trump and Joe Biden and Barack Obama.

0:03:46.800 --> 0:03:49.520
<v Speaker 1>Whereas the races for the House, or governor or Senate,

0:03:49.880 --> 0:03:53.120
<v Speaker 1>people are following those a lot less closely, voters might

0:03:53.160 --> 0:03:56.200
<v Speaker 1>not know anything about those candidates beyond their party affiliation.

0:03:57.800 --> 0:04:01.200
<v Speaker 1>For that target demographic of undecid I did swing voters.

0:04:01.640 --> 0:04:04.360
<v Speaker 1>Any new information provided by a tv AD about a

0:04:04.400 --> 0:04:07.120
<v Speaker 1>state or local race could swing them in the campaign's

0:04:07.120 --> 0:04:11.600
<v Speaker 1>desired direction. Many studies have been done on the effectiveness

0:04:11.600 --> 0:04:15.000
<v Speaker 1>of political tv ads and have shown mixed results. Some

0:04:15.040 --> 0:04:18.720
<v Speaker 1>studies determined that these ads have limited persuasive power. Other

0:04:18.760 --> 0:04:21.000
<v Speaker 1>studies have shown that it's the nature of the ads

0:04:21.040 --> 0:04:26.320
<v Speaker 1>that's the issue, positive versus negative. While Warshaw and colleagues

0:04:26.320 --> 0:04:28.640
<v Speaker 1>didn't look at the content of ads in their study,

0:04:29.120 --> 0:04:31.599
<v Speaker 1>he did point us to another study that showed all

0:04:31.680 --> 0:04:35.440
<v Speaker 1>types of ads a positive, negative, partisan nonpartisan seemed to

0:04:35.480 --> 0:04:39.520
<v Speaker 1>have the same small effect on voting outcomes. But in

0:04:39.560 --> 0:04:41.800
<v Speaker 1>the game of tv ads, the point is not to

0:04:41.880 --> 0:04:45.400
<v Speaker 1>convince a registered Democrat to vote for the Republican candidate.

0:04:45.640 --> 0:04:50.599
<v Speaker 1>It's to speak to the independent voters. Warshaw reiterates that

0:04:50.720 --> 0:04:53.760
<v Speaker 1>their research showed a very small effect of tv ads

0:04:53.760 --> 0:04:56.560
<v Speaker 1>on influencing people to vote for a certain candidate with

0:04:56.720 --> 0:04:59.920
<v Speaker 1>the largest plausible effect being one or two percentage point.

0:05:00.880 --> 0:05:03.520
<v Speaker 1>So if he was consulting on a campaign on how

0:05:03.560 --> 0:05:07.200
<v Speaker 1>to spend its money, he wouldn't automatically recommend pouring millions

0:05:07.200 --> 0:05:10.480
<v Speaker 1>of dollars into TV ads. The needle might simply not

0:05:10.560 --> 0:05:14.200
<v Speaker 1>sway enough. But as we saw in the last two

0:05:14.200 --> 0:05:17.400
<v Speaker 1>presidential races, there are plenty of contests that have been

0:05:17.440 --> 0:05:21.040
<v Speaker 1>decided by less than one percentage point a. Warshaw says

0:05:21.080 --> 0:05:23.440
<v Speaker 1>that this is also a lesson for political donors who

0:05:23.480 --> 0:05:26.039
<v Speaker 1>want the most bang for their buck. A donation to

0:05:26.040 --> 0:05:28.560
<v Speaker 1>a presidential campaign is a drop in the bucket, but

0:05:28.720 --> 0:05:31.479
<v Speaker 1>that same donation to a down ballot race could buy

0:05:31.520 --> 0:05:37.279
<v Speaker 1>more ads that actually influence the outcome. However, it's important

0:05:37.320 --> 0:05:41.279
<v Speaker 1>to note that research into past behaviors like Warshaws, was

0:05:41.360 --> 0:05:44.200
<v Speaker 1>looking at data from a time when most American voters

0:05:44.240 --> 0:05:48.320
<v Speaker 1>watched hours of cable television every day. Their study ended

0:05:48.360 --> 0:05:54.240
<v Speaker 1>with data from between one though a quarter of all

0:05:54.320 --> 0:05:57.600
<v Speaker 1>cable TV subscribers cut the cord and switched to streaming.

0:05:58.279 --> 0:06:01.320
<v Speaker 1>Projections say that fewer than half of all American households

0:06:01.360 --> 0:06:06.919
<v Speaker 1>will have cable TV subscriptions by digital ads like the

0:06:06.960 --> 0:06:09.560
<v Speaker 1>political ads you might see on Facebook or before a

0:06:09.600 --> 0:06:12.680
<v Speaker 1>YouTube video have been found to have an even smaller

0:06:12.720 --> 0:06:17.400
<v Speaker 1>effect on voters. But political advertisers are getting smarter now.

0:06:17.600 --> 0:06:21.159
<v Speaker 1>On ad supported streaming TV channels like Lifetime or Vice,

0:06:21.640 --> 0:06:25.360
<v Speaker 1>campaigns can run hyper targeted ads based on the subscribers

0:06:25.400 --> 0:06:29.440
<v Speaker 1>personal data. Because so many Internet platforms now collect and

0:06:29.480 --> 0:06:32.840
<v Speaker 1>sell our data, everything from what other things you watched,

0:06:33.040 --> 0:06:36.680
<v Speaker 1>read and buy, to demographics like age, gender and race

0:06:36.720 --> 0:06:40.200
<v Speaker 1>or ethnicity, to your voting record and the estimated value

0:06:40.200 --> 0:06:43.960
<v Speaker 1>of your home could all be influencing what ads your serve.

0:06:45.560 --> 0:06:48.240
<v Speaker 1>Just when you thought spooky season was over, You can

0:06:48.320 --> 0:06:56.520
<v Speaker 1>run from political ads, but you can't hide. Today's episode

0:06:56.560 --> 0:06:59.240
<v Speaker 1>is based on the article do campaign TV ads really

0:06:59.320 --> 0:07:01.960
<v Speaker 1>changed voters? Lines on hous to works dot com, written

0:07:02.000 --> 0:07:04.800
<v Speaker 1>by Dave Rouse Brainstuff it's production of by Heart Radio

0:07:04.880 --> 0:07:07.120
<v Speaker 1>in partnership with houst works dot com and it's produced

0:07:07.160 --> 0:07:10.040
<v Speaker 1>by Tyler Klang. Four more podcasts from my heart Radio,

0:07:10.240 --> 0:07:12.960
<v Speaker 1>visit the iHeart Radio app, Apple Podcasts, or wherever you

0:07:13.000 --> 0:07:14.320
<v Speaker 1>listen to your favorite shows