-
That graph, the first in the post now, shows the complexity.
It’d be interesting to see at constituency level but I feel any conclusion drawn would be very speculative, maybe I’d feel different on viewing the data.It’s wild how many decided on the day!
GoatandTricycle-
Yup, potentially complex when looking at 650 instances. Doesn't mean don't do it! And if there's an error in the methodology it can be pointed out and corrected. I'm not sure if speculative is the right word, but I get what you mean. But all survey/polling analysis has these limitations. We can never know the causes or things at the individual level. The social world is complex.
The key thing from my perspective is that I'm not looking for confirm anything! I want to see whether the claims being made (on both ends) are justified in the data.
It's the same thing. You can look at those as well. The question isn't about them though. But the data I want could be used to investigate questions around those parties as well.
Look at the link @greentricky just posted (thanks!). The second figure is essentially what I want to see. It shows movement/change from Tory to Reform and Labour. Cool - I also want to see this at a constituency level - that can offer some insights into the impact of that movement/change. The national level is interesting, but it doesn't answer the question: did this result in a split on the right that got Labour in, or was there enough movement from other parties to Labour to overcome any split.
That is, I want to see the second figure for each constituency not just the national level. I'm not trying to "confirm" anything.
I don't know what this means. I've already said that we don't know individual voter behaviour. Every analysis is like this.That doesn't mean we don't use aggregated data.