One of the issues for us in survey design is how to keep the survey down to a size that works for the people who are being interviewed or filling out the survey. It’s a fine line between achieving as much as possible from the survey – without annoying the respondent. As surveys can also affect how the respondent views a program or organisation, you need to avoid it being a PR disaster!
With web surveys, if it is too long, many people either won’t start it – or don’t complete it. With phone surveys, the irritation starts to show as people watch their clock and think of what else they could be doing. These responses are made worse if you appear to be asking the same thing more than once – or if the questions seem to be outside of the main focus of the survey.
Our way around it is to clearly define the purpose of the survey and design it around that purpose – what is it that we really need to know? We can then consign the ‘nice to know’ into the bin. Then the realism rule is brought in – what can we realistically expect from respondents in terms of depth and detail in 10-15 minutes maximum?
We also need to realise that most surveys are ‘top of the head’ responses. So, there is no use asking ‘how much did you actually spend on groceries last fortnight? A more realistic question would be ‘approximately how much is your weekly grocery bill? – and perhaps give them some options between certain ranges of dollars spent.
We work with our clients to ensure that the balance works – for them and their intended survey respondents.
Tags: questionnaire design
At this years Australasian Society’s conference, we submitted a poster – Survey Scrunches – what works and what doesn’t.