It’s election season again, so I thought it was an opportune time to talk about polling. By now I’m sure most of you have been exposed to the latest opinion data – on the telly, in the paper, over the internet and the radio. According to John Warhurst an established commentator of all things political – POLLS MATTER! Warhurst explains
“much of the flesh of an election year grows on a skeleton made up of opinion polls … in many ways, polls are markers on the campaign journey.”
While many will associate opinions polls with the political arena, polling has also found a home in the marketer’s toolbox.
Opinion polls are briefly defined as a survey of opinion from a particular sample. They are designed to represent the opinions of a select segment of the population, by asking a small number of people a set of questions and then extrapolating those answers to cover the larger group.
When used in marketing, opinions polls can be a cost efficient method of establishing and tracking trends over time, and have proven to be highly accurate in predicting outcomes when employed correctly. If used incorrectly, the potential for inaccuracy becomes great. For example, it would be careless to poll the elderly population over the internet, that is, unless you were only looking at the small segment of the elderly population who are internet users. It also follows that if you conduct a poll by fax, you can expect that the majority of respondents will be business people – I’m sure you get the point.
An agency that is serious about the accuracy of the data will be concerned with things like sampling error, non response bias, response bias, the wording of the question and also coverage bias. Failure to consider any one of these factors will cast serious doubts on the validity of the results, and acting on this unreliable data should be left to aficionados of ‘pin the tail on the donkey’.
The proliferation of the internet and now, SMS technologies, has given marketing a cost-effective platform from which polls can be conducted. In my opinion, this method of data collection is underutilised. Business operators can glean a great deal of qualified data from their customers with this method, with the potential for great rewards for consumers and even greater rewards for businesses.