8 Reasons Why Online Communities Fail – and, how we can help
At conferences, I’ve heard speakers say that you …Read more
There’s one big difference between an online survey respondent and a respondent who is asked questions by an interviewer. If you are the respondent, you are on your own when the survey is online. This makes a huge difference. No longer is there an interviewer there to cajole the respondent into finishing what might be an uninteresting subject matter. When conducting online surveys, the respondent can make their own mind up as to whether he completes the survey. Quitting the survey takes one click on that black cross in the top right hand corner of the screen.
If there’s one thing that annoys me about budget airlines, it’s their booking procedure. The fare looks fantastic until at the third screen, there are a stack of additions – airport charges, fuel surcharge charges, an extra cost for having luggage that might weigh more than a few feathers and, then, at the last minute a fee for your credit card payment. I call it cheating and I hate it. Most people do. So, when you are asking someone to complete your questionnaire online, let’s be honest. Tell them how long it will take and show a progress bar. It’s honest. If it is likely to take 15 minutes, say it will take 15 minutes.
Be clear and precise
It’s important to keep your questions as short as possible. If they are long, many people tend to speed read the question and may misunderstand what you are asking. If the words are unclear or ambiguous, they may refuse to answer, give a random answer or even give up. On rating scales, you must avoid long statements. If a statement is long, there will be a tendency not to read it carefully and answer as positively or negatively as other statements. This is often called ‘straight lining’ where a respondent might say everything is ‘very good’ because they generally like a product, for example.
Keep it interesting
It’s important to try and make your survey as interesting as possible. Choosing pictures rather than texts can make the experience more enjoyable. Similarly, if you have rating scales, respondents feel more engaged if they can use sliders or sort “cards” with statements on into piles interactively. It might sound counter intuitive, but although these techniques increase the length of time respondents take, research has shown that respondents are more likely to complete your survey if it is interesting and interactive.
Make sure your survey looks good on all devices
You don’t know whether a respondent will answer your survey on a PC, tablet or smartphone. It’s important to check that it looks good on each device. Questions with a long list of responses may not work on a smartphone as there will be a need to scroll down too far to find potential answers – you may even make respondents biased towards using the earlier responses. You might consider shortening question texts, response texts and the number of responses on a smaller device. Similarly, pictures may look great on a laptop or tablet, but too many images on a smartphone may be hard to see. Maybe, you need to use texts on a smartphone. A video may not look so good on a smartphone as it does on a bigger device. You might consider missing out questions relating to the video or, at least, analysing the data separately as different results may occur. It’s really important that the software you use supports these features.
Keep it short
There is nothing to stop an online respondent from departing as soon as a survey feels to be too long. If a survey is about a topic that respondents may be passionate, you may get away with a longer survey, but generally surveys should not take more than 10 minutes. And, don’t forget that progress bar. Respondents like to know the truth.
If it has to be longer….
Sometimes, clients or bosses can insist on questionnaires being longer despite your protestations. A good technique to employ in such cases is to break your questionnaire down into sections and rotate the sections, so that there is more likely to be an even number of people answering each section. Of course, you will need a software solution that can provide this.
Offer respondents the chance to take a longer survey
Again, if you want to conduct a longer survey, you might invite respondents to click on a link to a longer survey where you ask some more detailed questions. Permission interviewing works better than stretching the length of time you ask someone to do your survey.
Encourage, encourage, encourage
When I was at school, I remember a teacher telling me I was doing well when I was doing some art project – I was hopeless at art and found it boring. That timely encouragement helped me to complete, by my standards, a decent piece of work. As adults, we are no different. It has been shown that a message halfway through a survey thanking a respondent for taking part will work as an encouragement
Test your survey
Testing how your survey looks and trying it out on colleagues, friends etc. is important. And, don’t forget to try it on different devices. Instructions may not be clear on some devices. Generally, if instructions are long and complicated, you are probably asking the wrong question and a re-think is needed.
Using the right software
There are many survey platforms available. I think it is important that the software you use supports all of the above capabilities – not just some. It’s why we recommend Snap Surveys. It’s easy to use, but has the tools you need to carry out good quality research that respondents will find engaging. Why not try a free survey now and learn?