How fact-checking can positively impact online political campaigning
What happens when politicians are called out for sharing fake news? A project that analysed the accuracy of New Zealand political parties’ social media posts in the last election has had a marked impact on how politicians behave online—and plans are in place to continue the project to keep the spread of misinformation in check.
“Jack and I had been asked by the University’s Vice-Chancellor Grant Guilford to work on a project that fact-checked social media posts during the election,” explains Dr Krewel. “By coincidence, on the same day I’d been asked by colleagues overseas to participate in the DigiWorld project that compared digital campaign strategies around the world.
“Jack and I had a podcast where we announced we would do campaign monitoring and check all of the posts by political parties and their leaders in the four weeks leading up to the 2020 election. We released weekly preliminary results that covered things like which topics were dominant, how much negative campaigning we saw from the parties, and who was publishing fake news.”
Dr Krewel says the research team had a reasonably conservative definition of what constituted misinformation. “We didn’t try to pick up on instances where parties had different opinions, for example. We only measured when a story was completely made up and verifiably false, that we could fact check with independent sources,” she says.
“The number of instances of fake news wasn’t actually that high—in New Zealand it tended to be more ‘half-truths’, where most of the information is right but there’s something slightly dodgy about how it’s presented. For example, the National Party posted a poll result about which party people trusted more to recover the economy post-Covid. In the actual poll, National had a small lead over Labour—maybe two or three percent—but the bar graph they shared had Labour’s bar half the size of National’s. You couldn’t call it fake news, but it was certainly misleading.”
Dr Krewel says the uptake of the project by New Zealand media outlets was very high. “Our findings were really widely published, and we can show statistically that the political parties reacted to that by reducing the amount of misinformation they published over the course of the election campaign. It seems the parties and candidates had to fear reputational damage from being called out for negative campaigning and fake news—the stakes are probably higher in New Zealand compared with other countries because we’re a small country with a small media market.”
While the intention of the NZSMS was to monitor digital election campaigning, Dr Krewel never imagined the project would have such a positive effect on democracy. “Even the extremist niche parties reacted to what we were doing—they were spreading less misinformation and doing less negative campaigning when they knew we were watching them,” she says. “Usually, parties would get a ‘win’ from spreading fake or negative news, which is often much more likely to get them attention by being shared and commented on more than positive, accurate information would. But once they knew they were being watched and held to account, the ‘win’ turned into a downside for them.”
Dr Krewel says the idea behind the NZSMS was always to have a positive influence on the campaign. “The NZSMS of course had no impact on those parties who had already committed to a clean campaign in the lead-up to the election, which were the Labour party and the Green party who both had already promised New Zealanders a positive campaign and not using any fake-news long before we started our work,” she says. “But seeing a measurable decrease in misinformation from the other parties over the course of the campaign was a success, as far as we were concerned. We also hopefully had a positive impact on voters, too—if we called out political parties using fake news, then people might reconsider their own behaviour: not to vote for such parties, or not to trust everything they see on social media.”
The NZSMS’s results also reached the upper echelons of Facebook, with the social media giant removing Advance NZ from its platform after the party repeatedly spread misinformation. “They launched their own investigation into Advance NZ and found anti-vaccination content. That goes against Facebook’s terms of agreement, so they de-platformed Advance NZ three days before the election.”
Dr Krewel says she has no qualms about an independent piece of research having that kind of impact on an election. “For a second I was a little worried about a researcher having an influence on an ongoing campaign, with one party being completely removed from Facebook. But I have a clear conscience—this was a party that spread information that could potentially be life-threatening if people believed it. We did something good with our influence, so I can sleep easy at night!”
She says the NZSMS has shown there’s a definite need for fact-checking of political parties to be ongoing, and is in the process of setting up an Internet and Social Media Lab. “We found that parties’ reaction to being called out is unfortunately very short-lived—it’s basically a slap on the wrist where they’ll stop for a few days after being caught, and then start again. If you leave them unsupervised they go back to their old habits,” explains Dr Krewel. “It’s a good idea to have a social media lab doing constant fact-checking because politicians will quickly learn that they’re being watched by academics, and adjust their behaviour accordingly.”
Dr Krewel is working towards establishing the lab, and has support from across the University to get the funding together. “If I’m successful, this lab would be the first of its kind in New Zealand,” she says. “It will hopefully also work on commissioned research for government agencies or media outlets, and would be multidisciplinary, with students from a range of different majors benefitting by being involved in a real-world research project.”