Herald DigiPoll on MMP
The Herald DigiPoll asked 700 respondents about MMP:
- 35% prefer MMP
- 39% prefer FPP
- 10% prefer another system
- 16% do not know or won’t say
The Herald DigiPoll asked 700 respondents about MMP:
Polling Company: DigiPoll
Poll Method: Random Phone
Poll Size: 700 respondents, (3.8% maximum margin of error)
Dates: 15 September to 24 September 2008
Client: NZ Herald
Report: NZ Herald main story and breakdowns here and graphs here.
Party Support
Preferred PM
A UMR poll from 11 to 14 September found:
A UMR poll of 750 people on CERN’s Hadron Collider found
JafaPete blogs on US Research from Pew on whether or not respondents who have a cellphone and not a landline differ from those with landlines.
The answer in the US if yes they do, but the overall impact on findings is still relatively modest.
A Research NZ poll of 529 respondents on women’s equality was conducted between 2 September and 11 September 2008.
Polling Company: Nielsen
Poll Method: Random Phone
Poll Size: 1,130 total voters, 983 decided voters (3.3% maximum margin of error)
Dates: 11 to 17 August September 2008 (estimated)
Client: Fairfax Media
Party Support
Preferred PM
This follows on from the NZ First poll comparison. Again Roy Morgan has the Green Party higher than the otehr companies – but less so than with New Zealand First.
The graph above shows how NZ First have done in every public poll since the 2005 election.
There appears to be a significant difference between the level of support for NZ First in the Roy Morgan polls, as opposed to the five other polling companies. This is not to say Roy Morgan is necessarily wrong – just that there is some sort of systematic or methodological recurring reason why they show a higher level of support.
I don’t have available the full methodology used by Roy Morgan, so it is hard to speculate on why they get such different results. It might be linked to the fact they poll around 60 people a day and report in two week intervals, rather than the shorter time frames of the other polls.
Polling Company: Roy Morgan Research
Poll Method: Random Phone
Poll Size: 823 (3.5% maximum margin of error)
Dates: 01 September to 14 September 2008
Client: Self Published
Report: Roy Morgan Website
Party Support