One of the reasons for my love-hate attitude towards the state of modern British political science is the very limited approach taken to analysing and understanding tactical voting.
There are huge numbers of research papers looking at opinion poll data about how many people voted tactically in a particular election, in what sorts of seats and so on. Yet ask the collected authors when the first bar chart was used in a British election and none of them know. It’s a classic case of focusing on only one very particular methodological approach to an issue and completely ignoring others. In this case behave like a mathematician and don’t behave like an historian.
The reason I’m so confident none of the authors would know when the first one was used is that in fact no-one knows, which is why I’ve been on the hunt.
I had previously got as far back as 1984 and Chris Rennard’s Association of Liberal Councillors booklet, when he called them “block graphs” and included some examples from recent elections.
Now I’ve made it back to 1979 thanks to Richmond’s David Williams who has found this in his collection (see p.2):108898046-Liberal-Party-election-leaflet-Richmond-1979
Any takers for an earlier date? Let me know if you have some recollections or evidence.