Before looking at the results, it is vital to note that there are some important caveats about the survey. As a non-representative survey with a small sample, it must be stressed that the results should only be seen as indicative. I've added some key points about this at the end of this post, particularly with regard to the likely sample bias.

The mayoral election booklet

Of the 81 respondents to the survey, 93% said that they had received the mayoral booklet containing the election candidates of all 12 candidates and contextual information about the mayoral election and the electoral system being used.

Not only were the vast majority of those responding to the survey were conscious of having received the booklet, but 87% said that they had read at least some of it. Again, the potential impact of the bias in the sample must be underlined here (see below). But the value of the booklet as a mechanism for ensuring as many electors as possible receive basic information about the candidates' key policies seems apparent.

Indeed, one-fifth of those survey (21%) stated that the mayoral election booklet was their main source of information about the election campaign, representing the second most common answer after 'the internet' (36%). The local press was ranked third, with 11% of respondents citing it as their main source of information, and local radio fourth, with 9%.

It also appears that the election addresses of the 12 candidates contained in the booklet will have at least some influence on voting intentions. As the pie chart below shows, less than a third of those responding to the survey said it would have no influence on their choices about voting. Meanwhile, 43% said the information in the booklet would have a little influence on whether and/or how they would vote, 22% replied it would have a fair amount of influence and 6% said it would have a great deal.

Again, while caution needs to be exercised about what conclusions we draw from just 81 responses, my survey does suggest that the mayoral election booklet matters to this election campaign. This evidence suggests that the decision of the returning officer to require 2 candidates to change their election addresses prior to inclusion in the booklet should be seen as rather more than a storm in a tea cup.

If the information in mayoral election booklets is, as it seems, a key source of information for electors, then the need to clarify the rules on what rulings the returning officer can make about content becomes even more evident.

I'll be posting more results from the survey tomorrow.

Sample bias

While results were received from electors living in 15 of the 20 postcode districts of Liverpool, recruiting participants via e-mail, twitter and the web will almost certainly have biased the sample towards social classes A, B and C1.

The postcode districts for which no responses were received were predominately more deprived areas of the city, notably North Liverpool (L4, L5) and Speke (L24). There was a heavy bias in the sample towards residents of more affluent areas in South Liverpool, with 47% of responses coming from residents of 4 postcode districts: L15, L17, L18 and L19.

Despite this bias, it is important to note that election turnouts are considerably higher in the residential areas from which responses were collected. Hence, the survey has probably done a reasonable job of collecting responses from those most likely to vote in the mayoral elections.

Postscript: in addition to the caveats noted above, many thanks to @LplMayorWatch for pointing out that an internet survey may inevitably yield a larger proportion citing the internet as their main source of information!