What do we think of the Anthony Wells explanation?
YouGov staffer and operator of UKPollingReport, Anthony Wells, has put up a long post seeking to answer some of the points about the firm’s weightings that have appeared here in recent days.
The essence of his article is that there have been adjustments to deal with the new way it is doing the daily poll which do not involve members of its panel being asked to take part in a specific survey.
Anthony explains: ” Now people are sent a non-specific invite and, when they arrive at the YouGov system, they are allocated to whichever survey needs someone in their demographic group (for example, were I to arrive at the site the system would look at all the open surveys and see which oneâ€™s quotas were most in need of a 25-40, middle-class, Times reading man from the South East, and send me over there). What this means is that surveys donâ€™t get excess replies, respondents never get sent away empty handed, and that even very fast surveys get an even mix of fast and slow respondents, allowing accurate one day surveys.
However, because you canâ€™t tell who is going to respond to which survey, it also makes it more difficult to calculate the proportions of people to invite to get a sample that needs the least possible amount of weighting â€“ hence the slightly higher levels of weighting (though itâ€™s worth saying they remain relatively low compared to some of the weights needed for quasi-random samples)..”
The result, as far as I can see, is that we get the ridiculous situation that Labour seems to better when fewer respondents classified by the firm as being “Labour Loyal” take part. This is because, like in the example shown above, the responses from those that do are magnified in order for a poll to reach its “quota”.
Surely the answer is for YouGov to stop messing about with party ID weighting and use past vote weighting as employed by ICM, Populus, ComRes and Angus Reid?
UPDATE 1950 On reflection that previous point was wrong. It’s good that we have a range of pollsters all looking at the same issue but from different standpoints.