This is an in progress repo to estimate the effective sample size of political polls. It uses a pooling the polls model to estimate latent vote intention and the standard error of the polls.
The intent is to expand it to other countries and time periods, and hopefully use it to see if polls' margin of error has been increasing or decreasing over time.
This is being done out of self interest, but feel free to adapt the model as you see fit.
This repo makes use of the {britpol} R package (link here).
to come