*** Democracies Online Newswire - http://e-democracy.org/do *** *** *** *** Up to seven posts a week. To join over 2500 subscribers, *** *** e-mail <[EMAIL PROTECTED]>, in message: sub do-wire ***
UK Internet Poll Trial Research Report Published by ICM on behalf of the Guardian newspaper. [This is an apple/oranges comparision with the recently Pew survey that found conservatives more likely to fill out basic one questions online "snap" polls on web sites. - SLC] See: http://www.icmresearch.co.uk/reviews/2002/Internet-polling-paper- jan-03.htm Clips: 1. Introduction Considerable interest has been expressed in the possibility of conducting polls via the internet, and such interest has been heightened by the appearance in the UK of the internet polling company YouGov. Popular belief that opinion polls can accurately measure public opinion is based largely on their success in predicting the outcome of elections. Despite some notable failures (1992) for the most part polls in the UK have given a good guide to the eventual result. ICM’s telephone methodology, developed after 1992 with the full support of The Guardian’s editorial team, proved extremely accurate in particular in 1997 and 2001 as well as at several other elections in between. But neither ICM or The Guardian are complacent and recognise that past success is no guarantee of future accuracy. However before switching to a new polling methodology both organisations are concerned to ensure that any alternative they adopt is likely to be at least as successful as telephone polls. The accusation most easily levelled at on-line polls is that because only about half of the population have access to the internet their samples – and thus their results - cannot possibly be representative of the population as a whole. Nevertheless, YouGov have been successful in predicting the outcome of the last general election, the Conservative Leadership election and the first pop idol TV contest. So ICM, with the financial support of The Guardian, decided to test the efficacy of on-line polling. As a first step a series of questions have been put to a sample of 4,014 people interviewed between October 11th and 24th 2002 using ICM’s standard telephone poll methodology. In so doing we both identified those who have access to the internet and those who say they are willing to join an internet polling panel. The resulting data allow us to analyse and compare the characteristics and attitudes of those who are accessible via the internet and those who state that they are willing to join an internet panel with those of voters accessible to a telephone poll. If this test were to indicate that data collected from those willing to join an internet panel were capable of producing similar results to those of a telephone poll, we would have the confidence to go-ahead and set up an internet polling web-site in the knowledge that doing so would not undermine the reliability of ICM/Guardian polls. ... clip ... 8. Conclusions We have found that at present internet polls based on a recruited polling panel may not necessarily produce results that are representative of the population as a whole, even after very considerable weighting of the results has been undertaken or care exercised to ensure that those who are asked to complete an internet poll are demographically and politically representative of the whole population. Being on the internet reflects a difference of attitude towards life that is to a significant degree independent of socio-economic background. While it may be the case that internet polls conducted by YouGov may have hitherto avoided the potential pitfalls so far as voting intentions are concerned, this does not appear to be the case so far as least some social issues are concerned, most notably the Euro and the death penalty. It also appears highly likely that internet panellists are more politically interested and knowledgeable, and may perhaps be more inclined to take a left-wing stance on some issues too. As a result we would conclude that there is a risk that a switch towards internet polling could well undermined the reliability of ICM/Guardian polls, though this does not mean that the efficacy of this approach should not continue to be monitored. ^ ^ ^ ^ Steven L. Clift - W: http://www.publicus.net Minneapolis - - - E: [EMAIL PROTECTED] Minnesota - - - - - T: +1.612.822.8667 USA - - - - - - - ICQ: 13789183 *** Past Messages, Discussion http://e-democracy.org/do *** *** To subscribe, e-mail: [EMAIL PROTECTED] *** *** Message body: SUB DO-WIRE *** *** To UNSUBSCRIBE instead, write: UNSUB DO-WIRE *** *** Please forward this post to others and encourage *** *** them to subscribe to the free DO-WIRE service. *** *** Please send submissions to: [EMAIL PROTECTED] ***