Disclaimer – The views in this blog are in no way affiliated with any company, corporation or any individual and is the sole work of the author Ronnie Jones. I am a graduate from the University of Central Lancashire in the field of marketing and this blog is a general view on my thoughts on the subject of surveys.
A survey is a market research method to primarily gather quantitative data on a selected topic usually used for public opinion on products satisfaction to gather information on preferences, behaviour and factual information. The results can be used to create graphs with the goal to view current trends and to some extent predict new trends, companies who send new products on the market can use the tool to gather data quickly and on a mass scale to see how things are going.
I’m not a big fan of surveys – having conducted my own survey campaign on the subject of video game advertising as well as carrying out surveys with a large company; I can safely say that I found the answers given were heavily unreliable and a lot of time spent trying to sift through to try and make sense of it.
My main problem with the survey is how easily it is to influence the answers given even when a real attempt was made not to – the main areas of concern for this were
- Advertising the survey as ‘generic’ or ‘quick’ tended to make the answers generic and quick; meaning that the answers could be rushed and not thought out resulting in responses that weren’t necessarily true, nullifying the whole point of the research goal.
- The state of mind of the respondent can heavily influence the answer, catching the customer in a good mood will result in more positive answers, even if the customer was generally negative about the product or service anyway, again resulting in unreliable and skewed results. As an example, if you had a telephone survey and the customer was fairly happy about the product or service but was called during a meal time, doing work or with a troublesome child can affect the answers heavily, again resulting in warped answers.
- A more obvious conduit of unreliable answers are of the influence of the interviewee, if the respondent does or does not like the tone, accent, look or even the way the interview approaches the respondent this can result in unfavourable or favourable answers. This is one of the reasons where I would favour online research. Obviously this depends on the nature of the survey but that’s a topic for another time.
However, in defence of surveys their main goal is to get quick responses on a mass scale and it does that quickly, it is a cheap method so the costs are always low however the quality of the results are always low. I would always suggest using surveys as a base of a research campaign only but to tread carefully when taking any action on the basis of the research. In my experience I used online surveys to determine what the questions were going to be asked in my main research of group questionnaires where the proper time was given in the questions used and what answers where to be searched for.
Another suggestion would be to use a mathematical formula to determine what were anomalies, passive answers and unusable ones to be able have some type of spine for the data, otherwise a survey could end up taking the research in the wrong direction because of it’s high level of unreliability.
To conclude, have you ever seen a hair product campaign with ‘87% of women approve’ or something similar? Have you ever thought, what does ‘approve’ mean? Does that mean it’s good or does approve just mean ‘Yeah, I suppose this product works adequately’ whereas there is no real mention of quality or even value for money.