It is once again election year, and the frequency of opinion polls is starting to pick up. Yesterday NewsHub released the first Reid Research poll since before John Key stepped down as Prime Minister. Needless to say that did all they could to eek as much out of it as possible. However other elements of their reporting on it raise some questions.
In all of the graphics on social media, and all of the original stories that I saw there was no mention of sample size, nor the time during which the polling was carried out.
Now this isn’t the first time that NewsHub have failed to include these details. In the lead up to the Northland By-Election they did the same thing, and even when directly asked about it, still didn’t release the details:
It was only a day later, after another tweet, that we got some of the information:
So why is this an issue? Well, political opinion polls, and the role they do and should play in a democracy, is a contested area.
The research industry, and their industry body, are aware of this. To help address this issue the Research Association of New Zealand have published a New Zealand Political Polling Code. The introduction of this code states:
This code documents best practice guidelines for the conducting and reporting of political polls in New Zealand.
The code is binding on companies that are members of Research Association New Zealand and on researchers that are members of the Research Association New Zealand.
For each issue, the code details:
- Best practice for the market researcher conducting the poll
- Best practice for the market researcher in reporting results
- Best practice for the media in publishing results
The term “must” indicates a requirement, while the term “should” indicates recommended best practice.
The code covers a range of elements in the polling and reporting process. In this case the relevant sections are as follows:
So it seems that NewsHub are failing to follow the industries suggested best practice. Also, unlike OneNews and Colmar Brunton, who release the full report within minutes of the poll going to air, the Reid Research site still only has details of the last poll from July/August last year.
I tweeted about this last night:
And low and behold, this morning, the stories got updated with information about the sample size and timing:
This section appears to not have been added until 10am this morning, nearly 24 hours after the first story was published:
Is it too much for us to expect that the political editor of one of the two major news networks, and the reports under him, should be following the suggested best practice of the research industry body? Is it too much for us to expect that the research company will have made full details available 36 hours after stories based on the poll started appearing?
And while I am talking about polls, if political parties, or their youth wings, want to use the results of polls as part of their campaign, it would be a good idea for them to cite the specific poll they are using, and not wait to be asked deep into the comments: