Writing Good Surveys Is Harder Than You Think Pt 4
Part 4 of 4: Other considerations
Other considerations
Incentives
If you are unable to provide incentives to respondents (always a good way to increase response rates) you will need to make sure the survey is as short and sweet as possible, and distributed to as many people as possible to ensure quotas are reached.
Progress indicators
Progress indicators are very important for respondents to have a sense of where they are in the survey sequence. Good survey tools should offer one as an option.
Interactivity
Survey tools are becoming more and more interactive, enabling respondents to drag and drop and perform other interactions rather than just clicking. This makes the survey more engaging, particularly when it’s long. Consider this when choosing a survey tool.
Desktop and mobile
Make sure you consider what device(s) respondents will be completing the survey on. Most survey tools will enable your survey to be taken on a mobile, but it is best to check the implementation. You may be able to adjust the formatting and appearance of the survey to suit.
In addition, mobile surveys should probably be shorter than desktop versions, as respondents are using these devices in different ways.
Preview and test!
Always preview your survey and get people to test it before it goes out to the intended audience. There may be aspects you have overlooked.
Analysing and interpreting the data
It’s important to remember to cleanse the data before analysing it. Key things to look out for are when respondents have:
- skipped questions;
- entered nonsense responses;
- misunderstood the question;
- used the question as an opportunity to vent or meet an agenda;
- completed the survey too quickly compared to the average; and
- clicked responses in ‘patterns’ (revealing a lack of genuine engagement).
These responses need to be eliminated from your analysis as they either won’t make sense or will skew your data. A good survey tool will allow you to exclude them. Text responses are good indicators of bad data — another reason to include them in your survey.
One last thing
Remember your respondents are people! Don’t write a survey you wouldn’t be at least willing, if not happy, to complete yourself.
Props to the peeps at the now defunct Flow Interactive!
Missed something?
Part 1 — Making a plan
Part 2 — Structuring the survey
Part 3 — Constructing the questions
At SEEK we help people live more fulfilling and productive working lives and help organisations succeed. This provides us with a unique insight into the aspirations and needs of job seekers and hirers. Follow us at SEEK User Experience and SEEK Culture as we tackle the social and ethical challenges of producing software at scale that respects humans on an individual level.
