By Robert Tuomi
(WINDSOR, ON) – One of the most effective marketing tools is customer research. Large companies rely on polling for un-biased market information. Marketers use the data from various research initiatives, some of it on-going, to identify the true co-ordinates of where they stand in each of their product segments and also the impact that their competitors could have on their sales.
After the recent provincial election, pollsters were pretty much given a bad name. Inaccurate claims were made that they really had no clue about voter intentions. Much, if not all, was undeserved. And much of the criticism came from people who really know little about scientific-based research. One of them appears to be Keith Neuman. The Executive Director of the Environics Institute for Survey Research wrote a guest article published by the Globe and Mail on June 17. In it he seems to be exposing his own limited understanding of the value of survey research.
He holds a doctorate in social ecology which might explain some of his apparent confusion about the discipline particularly his claim that there is a growing, “difficulty in engaging representative samples of the public.”
Indeed, it could be quite the problem given, pollsters, including the famed American George Gallup, actually abandoned representative sampling in 1948 and then spent considerable time further refining the industry’s methods, a process that continues to this day. to meet the new challenges of a mobile world.
While some say the pollsters missed the results of the Ontario election, others aren’t so sure. Those who followed the website threehundredeight.com’s on-going review of various polls, right up to the election, saw that the pollsters were able to correctly identify the trend. They showed that after a rush of Progressive Conservative support voters began switching with the Liberals gaining the momentum needed for the majority win for Premier Wynne.
To understand how polling is done today is to understand why pollsters abandoned representative, or quota sampling. The process involved dissecting a population into its parts, such as men, women, or people who live in certain areas, and conducting personal interviews. It was costly and time consuming but what bothered the industry’s leaders is that there was no methodology to the selection.
Although representative sampling had served the industry well for over twenty years, starting in the early 1930s, its time had come. One of the reasons it initially worked was the homogeneity of the people being polled. As America diversified so too did its opinions. The way around this particular impediment to accuracy was probability or random sampling.
With this methodology, pollsters can attain very high levels of statistical accuracy. With market research now a staple of business, it should be no surprise that charlatans have entered the industry. There are no limits on who can or can’t pretend to be a pollster let alone a marketer.
The arrival of these pretenders caused Swedish polling commentator, Hans L. Zetterberg on his website, zetterberg.org, to lament that despite considerable progress in making polling methods better, unfortunately, “some 50 years later, we see new pollsters emerging, often in market research firms, who do not seem aware of these solutions. And many journalists who write up or present polling results are unaware of the problems defined in 1948 and their solutions.”
In short, polling is critical to understanding where a politician or a business is and where it is going. Any problems the industry is accused of having may be as much about polling as it is about those doing the polling. Because it is so critical it should be turned over to people who have the appropriate training, education and experience. Otherwise the effort could be wasted.
If you have a question about marketing, you’ve come to the right place. Let us know and we’ll give you an answer to help your business.