@lottehargrave
Lecturer in Quantitative Political Science at The University of Manchester. Research gender, political behaviour, elections, and representation https://www.lottehargrave.com/ Non-academically, lover of: - Hiking 🥾 - Wild swimming 🌊 - Pub going 🍺
The Greens came second in 40 Labour-held seats at the 2024 general election. These were all seats Labour won. This should be ringing alarm bells for Labour.
Gorton & Denton was only one contest, and a byelection at that, but it highlights a dynamic that Labour may want to pay closer attention to.
Labour has responded to its drop in the polls by positioning itself against Reform, particularly on issues like immigration.
But if British politics is increasingly developing into left–right blocs, Labour may be unwise to ignore the threat from its own side of the ideological spectrum.
One possible explanation is that Reform’s rise may be mobilising both its supporters & opponents.
In both contests, voters backed a party to Labour’s left that appeared to be the strongest local challenger to Reform.
Rather than staying home, some voters may be turning out to prevent a Reform win.
Turnout was also notable: byelections normally see lower turnout, but we didn’t see this here.
Turnout in the seat in 2024 was 47.8%, and it fell by just 0.3 percentage points in the byelection.
We saw something similar in the Caerphilly byelection last year, where turnout actually increased.
Women were more likely to be undecided: 36% of women didn't know how they'd vote vs 18% of men.
That matters because Reform tends to perform better among men, while the Greens perform better among women. If many of those deciding late were women, that could help explain the late swing to the Greens
A key feature of the campaign was the high number of undecided voters.
In the week of polling day, 31% of voters still hadn't made up their minds. A figure higher than the support for any party.
For comparison, ~12% were undecided two days before the 2024 GE, which was already considered high.
The result itself was dramatic: the Greens overturned a large Labour majority to win the seat, pushing Labour into third place behind Reform.
We saw the overturning of Labour's 70th largest majority, and one of only 71 seats where the party won an outright majority in 2024, out of their 411 seats.
New piece from w/ @hannahbunting.bsky.social & @jess-smith.bsky.social on the Gorton & Denton byelection, and what it might tell us about British politics.
We cover turnout, late-deciding voters, gender gaps & the continuing fragmentation of the party system 🧵
theconversation.com/late-decider...
Green MPs Hannah Spencer, Ellie Chowns, and Carla Denyer standing in the House of Commons chamber
Hannah Spencer being sworn in as an MP
Earlier today, the new Member of Parliament for Gorton and Denton, Hannah Spencer was sworn in.
Well, no, they’re not just as wrong. The bigger error is not seeing that “trying to win over voters from a party on the opposite end of the spectrum, whose values your core voters oppose” is not same as “chasing voters from party on same side ideologically whose values your core voters like.”
One factor could also be how pollsters are handling undecided voters. It's a well-known fact that women are more likely to report DK (especially far from election day). If we think women are more likely to vote Green & some pollsters use squeezed VI in a follow-up, that could partially explain it.
Well, now she’s made it in the door, it’s time to close it for everyone else
And with that citizenship, a voter in yesterday’s by-election too!
Poor Matt Goodwin. All he ever wanted was to be a hard working MP/academic/tv presenter/commentator/writer/pollster/wealth adviser/women’s health guru and you all ripped that away from him. A shame
Check out the post below by @philipjcowley.bsky.social, who’s written a great summary of ongoing work by me, @hannahbunting.bsky.social, @cerifowler.bsky.social, @jess-smith.bsky.social, and @annasanders.bsky.social on gender gaps in “don’t know” responses👇
New Swingometer! The Gorton and Denton By-election: A tale of Two Manchesters
open.substack.com/pub/swingome...
Or the parking on the pavement crisis
This week, an X user created an AI video of me being chloroformed and prepared for rape.
Victims are told it's not real and they're getting offended for no reason.
But nudification tools are weapons of sexual assault, and the harm they cause is real 👇
jessasato.substack.com/p/ai-nudific...
Share a photograph of a mountain that you’ve taken
(Pico do Arieiro, Madeira - a place I think about almost daily)
Lab to DK is a group twice as large as Lab to Ref (and losses to DK are often easiest to win back)
Lab to Grn/LD is a group three times as large as Lab to Ref (and on multiple metrics this group looks more open to returning to Labour)
🚨New article in @electoralstudies.bsky.social🚨
New method using open-text survey, parliament speech analysis & conjoint experiment to detect policies/issues where all of:
1) bottom-up public demand
2) elites are neglecting it
3) would motivate vote choice if party adopt it
tinyurl.com/44ryyybc
🧵
There is one week left to apply for @uclspp.bsky.social's new Lecturer in British and Comparative Politics role!
Importantly, the successful candidate will also join our senior team and contribute to our research and impact activities.
Applicants must have, or be near to finishing, a PhD.
Apply 👇
What do ex-Labour Reform supporters believe in?
Data shows estimated difference in support between “Labour Loyalists” and “Labour to Reform Switchers”
By @lottehargrave.bsky.social.
📄 Data documentation is available on the OSF here:
lnkd.in/ejbhMsQV
More from us soon on this! 👀
In the meantime, we’re excited to share our data documentation for the hand coding we’ve done of Waves 1-29 of the British Election Study Internet Panel.
We have ~250 million observations to play with 📊 and we’re applying for funding to extend this work to other major social science surveys.
Right now, we’re running machine-learning models to identify who answers “don’t know”, to which questions, and when.
This matters because the standard practice in survey research among academics and pollsters is to “drop the don’t knows” before analysis.
When we do that, we are systematically excluding many more women than men from our data.
We’ve confirmed that women are around twice as likely as men to select “don’t know”.
Crucially, this varies by what questions are asking about, and by the choices we make as survey designers!
🧵 Thread on new research on gender gaps in “don’t know” responses
Some of you will know that my stellar coauthors @hannahbunting.bsky.social @cerifowler.bsky.social @jess-smith.bsky.social @annasanders.bsky.social & I have been working on a large-scale project on “don’t know” responses in surveys.
🚨 NEW BLOG
Labour have won every election in Wales for 100 years, but they are on track to (badly) lose the 2026 Senedd election - why?
@jaclarner.bsky.social and I have looked at new data, which shows how support is shifting within (not between) Wales's blocs!
blogs.cardiff.ac.uk/thinking-wal...