In-built skewing of election poll methodology

To post in this sub-forum you must have made 100 posts or have Trust status or have completed our ID Verification

Forums › Politics › In-built skewing of election poll methodology

Geektastic

18009 posts

Uber Geek
+1 received by user: 8465

Trusted

Lifetime subscriber

#152068 15-Sep-2014 08:49

I noted that several polls get their samples from households with landlines.

I think people - especially those of a younger age group - will increasingly dispense with landlines. For example we've not had one for almost 2 years and I doubt we are unique in that.

Is it time for that changing telecoms landscape to be reflected in polling?

Behodar

11092 posts

Uber Geek
+1 received by user: 6070

Trusted

Lifetime subscriber

#1128765 15-Sep-2014 08:51

I haven't had a landline since moving out of my parents' place last decade :)

tardtasticx

3084 posts

Uber Geek
+1 received by user: 483

#1128796 15-Sep-2014 09:13

Not a single one of my friends at uni own a landline. We all get stupid amounts of minutes included in our mobile plans so there's no need to even consider it.

I think it'll only spread more as generations grow up and the new lot move out of home and realise how useless that thing is.

Batman

Mad Scientist
30012 posts

Uber Geek
+1 received by user: 6217

Trusted

Lifetime subscriber

#1128803 15-Sep-2014 09:24

Until you have a natural disaster that is, but only if you have people to track down, otherwise there still is no use.

YadaMe

110 posts

Master Geek
+1 received by user: 44

#1128811 15-Sep-2014 09:35

joker97: Until you have a natural disaster that is, but only if you have people to track down, otherwise there still is no use.

Except in Christchurch the best comms was text message. Both landline and mobile voice were out

NZtechfreak

4649 posts

Uber Geek
+1 received by user: 461

Trusted

#1128829 15-Sep-2014 09:57

This isn't news, this has been a criticism leveled at phone polls for some time.

Twitter: @nztechfreak
Blogs: HeadphoNZ.org

Geektastic

18009 posts

Uber Geek
+1 received by user: 8465

Trusted

Lifetime subscriber

#1128954 15-Sep-2014 12:08

NZtechfreak: This isn't news, this has been a criticism leveled at phone polls for some time.

I'm not claiming it is news.

Just wondering why no change is happening when we are so poll obsessed. Personally I would ban them for 10 weeks before the election but since we have them, we ought to be making noises that they are potentially ignoring a swath of voters simply because of methodology.

Geekzone

Want to support Geekzone and browse the site without the ads? Subscribe to Geekzone now (monthly, annual and lifetime options).

wasabi2k

2102 posts

Uber Geek
+1 received by user: 860

#1128967 15-Sep-2014 12:28

given NZ's appalling voter apathy, particularly among younger voters - it is probably entirely representative.

Yes, that was tongue in cheek. No polling methodology is 100% and they all have bias. Informed people take any poll with a grain of salt.

They publicize the results of ONLINE polls with little to no quality control ffs.

BarTender

3629 posts

Uber Geek
+1 received by user: 2572

ID Verified

Trusted

Lifetime subscriber

#1128992 15-Sep-2014 13:25

Geektastic:
NZtechfreak: This isn't news, this has been a criticism leveled at phone polls for some time.

I'm not claiming it is news.

Just wondering why no change is happening when we are so poll obsessed. Personally I would ban them for 10 weeks before the election but since we have them, we ought to be making noises that they are potentially ignoring a swath of voters simply because of methodology.

I would say the main issue is there is no easy register of phone numbers to use for mobiles and the number ranges are that much larger. Plus there is no geographical constraint on mobile so it would be hard to poll under 25s in Auckland.

So to get a representative result dialing a whole number range may be quite costly on polling companies and yield a low hit rate.

Would be interesting to see David Farrahs view on it since he is the chief pollsters for National.

afe66

3181 posts

Uber Geek
+1 received by user: 1678

Lifetime subscriber

#1128994 15-Sep-2014 13:33

And include the margin of error in a way the Joe public actually understands.

Ie
Conservative party 2.6 to 7.3% rather than 4.6% with margin of error of 3.1% in tiny writing. As this would show party A increasing its support by 1% and spending 10 minutes discussing this on news rather than the actual polices, also reinforces these changes might not actually mean anything.

I also agree with another poster that we should ban polls for 12 weeks before the election.

alasta

6888 posts

Uber Geek
+1 received by user: 3362

Trusted

Subscriber

#1129449 15-Sep-2014 21:04

BarTender: I would say the main issue is there is no easy register of phone numbers to use for mobiles...

That doesn't seem to be a problem for people who ring me up to sell me life insurance, among many other things.