The usual pattern of responses to our monthly survey was for about 500 or so Party member replies to arrive on day one.  Another 200 or so were then received on day two.  Enough new responses then appeared on further days usually to take the number of replies above 800.  The survey itself ran via a link on the site for the best part of a week.

This was before we decided to improve the survey’s security against any attempts to skew it, remove any link to it from the site entirely, and instead e-mail it out to approximately 3500 Party member respondents.  Their details were obtained from the very substantial returns received from recent surveys, and they had not objected to becoming part of a new panel.

We will be publishing the first results obtained since the change this week, but can today provide an overview of the responses we receieved.  The email was sent to 3,556 people.  At least 1830 (57 per cent) opened it (although the real figure could be higher, as some servers don’t provide this feedback data).  1343 of these (77 per cent) clicked the survey link, of which 1234 actually replied.

1379 people completed the survey altogether.  This figure takes into account those to whom we e-mailed the survey separately because their e-mail provider chucked it into spam, plus Party members who wrote asking to be included, and were able to provide proof of membership – an up-to-date card, say.

The response pattern was different.  1092 replies to the original e-mail were arrived on day one, with only relatively small numbers coming in on days two, three and four (104, 30 and eight respectively).  Respondents evidently feel more inclined to reply quickly to an e-mail, and feel that they can take their time to click on a link.

As we wrote last month, we hoped to get at least 750 replies.  So 1379 is what John Major would call “very satisfactory”.  Opinion polls of voter intention regularly sample about a thousand voters out of some 45 million.  The new survey obtained over a thousand out of about 150,000.

This raises the question of how we can be sure that the 1379 voters are party members, and whether their view is representative, since the survey is not a scientific poll.  The long and short of it is that we cannot guarantee that there is no infiltration.  But there is no reason to believe that a significant proportion of those who have declare themselves to be Party members are not so.

Further to that point, it’s worth noting that last month’s Next Party Leader findings replicated very closely a recent YouGov poll of Party members on the same subject.  The survey has tended to produce responses that are nearer the mainstream than those of our non-Party member readers.  And over the years it has been very consistent – see here, here and here, for example.

We hope to return to obtaining and publishing responses from our non-Party member readers, but want for the moment to concentrate on continuing to improve the Party member findings.  Are the findings of the new survey out of line with the old one?  Readers can begin to find out the answer tomorrow.