Do you have number crunching skills we can call on? Right here is a spreadsheet with what demographic data is available for the Tomorrow's Europe Deliberative Poll. You can login as readers (@opendemocracy.net), password: readers . You can also download the Excel version from here.
Some of our questions:
- what are the entire EU proportions for these demographics?
- what is the confidence level that the organisers have used to determine confidence?
- are there any borderline cases?
- why have these demographic categories been chosen for control?
- how badly under-representative is the nationality sample?
- can we draw any inferences about the impact of nationality on belief given the sample size?
- what can we say, overal, about the degree of representativeness of the sample?
- what factors would you have controlled for if you had been designing the experiment?
Professor Fishkin, one of the masterminds of the deliberative polling method, has set out three criteria by which to judge the success of the Tomorrow's Europe deliberative poll. The first, and arguably most important of these is "Is it representative?"
Much of this will naturally come down to opinion, so in an utterly unscientific test, let me know what you think. Check out the spreadsheets, play with the numbers, and let me know the results of your findings via james.clivematthews [at] opendemocracy.net - assistance from statisticians especially welcome!
