After four weeks of the 2026 New Zealand Social Media Study, researchers at the Internet, Social Media and Politics Research Lab, led Dr Mona Krewel , have analysed 2,222 Facebook posts by political parties and their leaders. The posts were published from 5 October to 1 November.
Latest analysis investigates how frequently parties engage in negative campaigning, who is attacking who, whether attacks have intensified as election day nears, and if the 2026 campaign has been an exceptionally negative one.
Who has been the most negative Facebook campaigner in 2026?
Accusations of negative campaigning have become a recurring feature of political debate. But which parties have actually spent the largest share of their Facebook campaigns attacking their political opponents?
Across the past four weeks of the campaign, the National Party has emerged as the most negative campaigner, with 65.8 percent (371) of its Facebook posts containing attacks against political opponents, says Dr Krewel.
In second place is NZ Outdoors & Freedom (52 percent or 152 posts), followed by Labour (51.4 percent or 214 posts).
Which party is the least negative? Te Pāti Māori, which engaged in negative campaigning in just 15.8 percent (29) of posts.
Figure 1: Negative campaigning by party and leaders (% of party + leader posts)
Is the campaign more heated as we approach election day?
As election day draws closer, it is plausible that political competition intensifies and parties become increasingly willing to attack their opponents.
However, analysis of Facebook posts shows no consistent pattern of negative campaigning among parliamentary parties.
“We see different strategies, with some parties ramping up their negative rhetoric while others appear to shift towards more positive messaging in the final days of the campaign,” says Dr Krewel.
“National shows a consistently high degree of negativity throughout. In contrast, Labour shows a more noticeable increase towards the end of the campaign. ACT and NZ First have become less negative, while the Greens show a modest increase in negativity.”
Figure 2: Negative campaigning posts by parliamentary parties (%, 3-day moving average)
Among non-parliamentary parties, the picture is more mixed, with marked fluctuations in negative sentiment throughout the campaign.
“The sharp fluctuations among smaller parties partly reflect their lower posting frequencies. When a party publishes only a handful of posts, even one or two negative posts can produce substantial changes in the percentage.
“What stands out though is that The Opportunity Party (TOP) maintains comparatively low levels of negative sentiment, with several periods in which none of its posts contain negative messaging. It seems to be eager to get the message across that it is willing to work with everyone post-election,” says Dr Krewel.
Who is in the line of fire?
Negative campaigning is not unusual in democratic elections. Criticising an opponent's policies or a government’s record is a legitimate part of political competition.
“What matters, however, is whether attacks focus on substantive political disagreements or seek to undermine opponents on a more personal level. This makes it worthwhile to look at who gets attacked and how personalised these attacks are,” says Dr Krewel.
The findings show “ideological distance” is a deciding factor when parties choose who to attack.
“This is a well-established finding in political science research. The greater the ideological distance between parties, the more likely they are to attack each other. National primarily directs its attacks at the current parliamentary opposition parties and their leaders. None of its posts attack NZ First, Winston Peters, or David Seymour, and only one post (0.27 percent) in four weeks attacked the ACT Party,” she says.
Figure 3: Attack targets of the National Party and its leader (% of negative posts)
In a similar vein, Labour targets the parties currently in government and their leaders.
Figure 4: Attack targets of the Labour Party and its leader (% of negative posts)
Both major parties devote almost no attention to parties outside Parliament, most of which have limited electoral prospects and are therefore unlikely to influence the election outcome, Dr Krewel says.
“However, one notable exception is TOP, which features in 5.4 percent (20) of National's negative Facebook posts, as well as Qiulae Wong, who gets attacked by National in 1.4 percent (5) of negative posts. One possible explanation is that National is trying to prevent TOP from crossing the five percent threshold and entering Parliament.”
Dr Krewel says National's attacks on TOP, together with its negative campaigning against the Green Party’s co-leader Chlöe Swarbrick, also point to a more unusual pattern.
“While major parties occasionally attack smaller parties, their primary targets are usually their main competitors in the opposing political bloc. In New Zealand, we would therefore expect National to concentrate its attacks on Labour and Chris Hipkins, while Labour's natural go-to targets would be National and Christopher Luxon.
“For Labour, this is what we found. It almost exclusively focused on Luxon (59.3 percent or 127 posts) and National (47.7 percent or 102 posts), and only to a limited extent attacked Seymour (3.3 percent or 7 posts), ACT (1.4 percent or 3 posts), NZ First (0.93 percent or 2 posts), and Winston Peters (0.5 percent or 1 post).”
The picture for National, however, is a different one from what would normally be expected.
“The considerable attention National has devoted to Chlöe Swarbrick stands out. Almost 27 percent (100) of its posts target her. It is unusual for the leader of a smaller party to attract this much negative campaigning from one of the two major parties. Swarbrick has faced even more attacks from National than Labour leader Chris Hipkins, who’s targeted in 23.5 percent (87) of posts.
“We can only speculate about why National has devoted so much attention to Swarbrick. One possible explanation is that Hipkins is not a particularly polarising figure. However, another is that National has chosen to portray Swarbrick as a political ‘bogeyman’ and use this scenario to stoke fears among middle-class and more conservative voters who might otherwise be less concerned about a change in government.”
Dr Krewel has some doubts about whether National's attack strategy against Swarbrick is working, given posts targeting the Greens co-leader attract, on average, more angry emoji reactions than posts targeting Chris Hipkins.
“This is particularly striking because politicians' Facebook posts typically attract predominantly positive reactions, such as likes and loves, as their audiences largely consist of supporters who follow them to express their political allegiance, while those who dislike them are generally less likely to follow their accounts.
“The unusually high number of angry reactions could therefore be a sign these attacks are not being received as positively as National might have hoped, even among its own supporters. It’s possible the angry reactions may come from supporters of the left bloc who intentionally visit National’s and Luxon’s pages. However, given this requires more user effort, this seems the less likely explanation,” she says.
Who is doing the dirty work?
Political science research across several countries has identified a strategic division of work in negative campaigning. Parties often leave the attacks on political opponents to their general party accounts or lower-profile politicians, while party leaders maintain a more positive public image.
“Parties strategically delegate the dirty work to their party organisations or less prominent politicians, allowing their leaders to stay above the fray and focus on promoting a more positive image of themselves,” says Dr Krewel.
However, posts targeting Swarbrick aren’t sticking to this pattern.
About 31 percent (18) of the negative posts published on Luxon's Facebook account contain attacks against the Greens' co-leader, compared with approximately 26 percent (82 posts) on National's official account, she says.
“While the difference between the two accounts is relatively small, it is nevertheless striking that Luxon personally engages in negative campaigning against Swarbrick even more frequently than his party's official account. Rather than distancing himself from the attacks to protect his personal image, Luxon gets his hands dirty. This runs counter to what we would typically expect.”
Figure 5: Negative campaigning against Swarbrick by National and Luxon (% of negative posts)
Is 2026 more negative than previous elections?
There has been plenty of debate about whether the 2026 election campaign has been particularly dirty and negative. And analysis shows negative campaigning is increasing.
The researchers compared the proportion of negative Facebook posts published by political parties during the 2020, 2023, and 2026 election campaigns. To ensure comparability, the analysis was restricted to the six parties (Labour, National, Greens, NZ First, ACT and Te Pāti Māori) currently in Parliament that contested all three elections.
“The results reveal a striking trend: negative campaigning has increased substantially since we started tracking Facebook posts in 2020,” says Dr Krewel.
In the 2020 election, just 12.8 percent (115) of posts published by the six parties contained negative campaign messages. This proportion rose to approximately 41 percent (430 posts) in 2023 and reached 55 percent (591 posts) in 2026.
“The data suggest the perception of an exceptionally negative campaign is justified. The proportion of Facebook posts containing negative messages has increased since 2020. In fact, more than half of all posts published by the six parties in our comparison now contain some form of negative campaigning,” she says.
Figure 6: Negative campaigning in 2020, 2023, and 2026
After the election, Dr Krewel and her team will publish further analysis, wrapping up their findings from across the campaign and examining how political parties’ social media strategies in 2026 compare with those of previous New Zealand elections.
About the New Zealand Social Media Study
The New Zealand Social Media Study was run in both the 2020 and 2023 elections, and is running again this year. The study tracks Facebook posts by political parties and party leaders in the five weeks leading up to the general election. The study sources Facebook data from the Meta Content Library.
Nine parties are included in the 2026 study:
- six parties currently in parliament—National, Labour, Green Party, ACT, New Zealand First (NZF), and Te Pāti Māori (TPM)
- three parties outside parliament—The Opportunity Party (TOP), Conservative Party, and NZ Outdoors & Freedom Party.
This year, the research team is using a hybrid approach to analyse Facebooks posts. The approach combines artificial intelligence (AI) for most coding tasks with independent validation by human coders. The main AI model used is Mistral Large 4, supported by Xiaomi MiMo-V2.6-Flash and Qwen3-Max.
The research team provided the AI model with detailed instructions based on a coding framework. This framework, used in the 2020 and 2023 studies for manual content analysis of Facebook posts, defines what to look for and how to classify it. Analysis considers written text as well as images and videos.
All AI classifications are automatically checked for consistency and compliance with the coding rules. Where claims need to be fact-checked, additional external sources are consulted and evaluated, and human data annotators further validate the LLM classifications.