Both the Herald and Stuff websites are currently running so-called 'opinion polls' on the Judith Collins situation. I'm writing here not to comment on the Justice Minister's predicament, but on the complete and utter mockery that such polls make of decent research.
Media websites like those polls because they generate interest, but they have no value whatsoever in terms of gauging public opinion. What's more, they give proper polls a bad name. A lot of people don't seem to be able to tell the difference between proper polls which at least attempt to write fair and balanced questions and to sample a broad cross-section of the population, and dodgy polls that do neither. Some in the media don't seem to be able to tell the difference either, and there have been several cases recently of newspaper websites claiming that their website polls were in some way useful as a gauge of public opinion.
It's not just website surveys either. In election years, we've often seen vox-pop type street surveys passed off as legitimate gauges of public opinion. I thought it might be fun to go through some particularly ridiculous examples of straight-faced reporting of garbage polls.