Giving surveys a bad name
Humanity in general is genetically predisposed to not reply to surveys, but in some people a genetic variation makes them positively want to answer surveys. At least, that's the conclusions reached by researchers who... carried out a survey.
Lest that seems like flawed logic, the survey carried out by scientists in the USA and Singapore was carefully planned. They didn't simply ask people whether or not they responded to surveys, but instead asked a thousand sets of twins to take a lifestyle survey and measured the response rate.
In doing so they found that the behaviour of one identical twin when completing the survey was a very good indicator of the behaviour of the other twin, whereas in non-identical twins there was no such strong correlation. This suggests that it is genetics rather than upbringing which determines people's likelihood to respond to surveys, genotype rather than phenotype.
This was an example of a well-designed and well-run survey. Sadly the internet is full of surveys these days, often very badly designed, inadvertently full of loaded and leading questions, and with no understanding of how to construct a sample so that the results are meaningful. In the same way that many people think that a word processor turns you into a great author, so too do people believe that a bit of survey software turns you into a qualified researcher and statistician.
The prevalence of surveys also creates opportunities for spammers who now send out adverts in the guise of surveys and claim it is legitimate market research. In the UK, telemarketing companies also frequently claim to be doing "telephone research" when they cold call you and say that because it is research, it is exempt from the telemarketing industry's self-regulation and the requirements of Ofcom's Telephone Preference Service.
Data harvesting in the guise of survey is also growing increasingly common, especially scams which promise that respondents will be entered into a prize draw or receive discount coupons for answering a few questions. McDonalds frequently features in these surveys which are designed purely to collect names, addresses and credit card details, but the food chain giant has no connection with these surveys at all.
According to a study conducted by F-Secure, a recent online scam based around a McDonalds survey on a social networking site attracted 32,000 visitors, with 12,000 of them completing the survey and handing over personal data. A security adviser for F-Secure said that because conversion rates are so much better than for spam, you can be sure fake surveys won't be going away any time soon.
Readers who require assistance conducting genuine surveys which are properly designed, properly analysed, and operated according to Market Research Society guidelines should feel free to contact us, or look at our web survey site:
29th September 2010
This article comes from the SKILLZONE email newsletter, published monthly since January 2008, and covering topics related to technology and the internet. All articles and artwork in the SKILLZONE newsletter are orignal content.