The one thing you can bet on about any online survey is that some people on the list will not fill out the questionnaire. It’s actually a problem for all surveys though, not just the online kind. Some people will not answer survey phone calls or letters, and others will not have the time to take a face-to-face interview. It’s just one of the hazards of the business. But it can still get quite frustrating.
Actually, it can get worse than frustrating. If most of the people who fail to respond to a survey share a particular view, the survey will return misleading conclusions. Non-response bias is the phrase we use to describe it. If, for example, we would conduct a survey asking our readers whether this article is informative or not, and most of the people who think it’s a lousy article decided to be polite and not respond, the final data might show that it is the best article ever written. That would be lovely for the writer’s ego – thank you very much everyone - and a strong basis for demanding a pay rise, but it would also mean more lousy articles for everyone.
Now imagine if the survey was about the market readiness for a new multi-million-product. If the majority of those who fail to respond feel the proposed new product is a no-go, that particular unfortunate company would end up wasting millions on a product destined to be dead on arrival. It would make the survey worse than not having a survey at all.
So, how can you avoid non-responsive bias?
Tip #1: Be Very Careful With Your Sampling
Poor sampling is the first point at which non-response bias is built in. Targeting the wrong respondents or reaching out to more people than necessary increases the chances of people not responding. Of course, sampling is a complex affair and can be quite technical. In brief, good sampling minimizes the overall number of respondents and focuses on potential respondents who are most likely to respond.
Tip #2: Keep the Questionnaire Brief
Sometimes, the length of a questionnaire is all that discourages a respondent from responding. People may be interested in the survey and even eager to participate, but if it looks like it will take forever, they will understandably skip it.
Tip #3: Make The Questionnaire Good Looking
This is a point that is often overlooked. Being beautiful or handsome is a strategic advantage, even if you are just a humble online survey. The type of fonts, the color scheme, and the layout either makes the survey look ‘respectable’ or not. The company doing the survey and how it presents itself also projects the ‘seriousness’ of the survey.
Tip #4: Give Respondents Some Incentives
This is an increasingly common technique, although it is still somewhat contested among expert surveyors. The incentive may be straight-up cash offers, discounts on online purchases, gift certificates from major online retailers such as Amazon, or access to premium online content.
Tip #5: Analyze the Details of Those Who Responded
Analyze the personal details - age, area of residence, income, and education levels – of those who responded against those who did not. Expert analysis of the results may indicate what perspective the non-respondents hold, which can then be factored into the survey data to limit non-response bias.
Tip #6: Compare Fast Responses Against Slow Ones
Again, based on personal details, compare responses from those who responded quickly against those that came back later. Research survey studies suggest that late respondents often reflect the views of those who did not respond.
Tip #7: Follow Up On the Non-respondents
A follow-up survey, either still online or by phone, usually surprises respondents enough to elicit a response. The result may be all that is required to solve the non-response bias.
Now that you know how you can increase your response rate for your online survey, try making one now and see for yourself what kind of results you will get!
P.S. If you liked this article, you might also like The Effect of Sample Size On a Survey's Results, How Knowing Your Respondents Can Help Improve Your Survey Results, and How To Minimize Sampling Errors In Online Surveys.
Mar 28, 2014