A survey can be a great way of getting information from a target population to help with your research. If you plan your survey well, you can get a lot of rich and valuable data as a representative sample of your population. Planning a good survey can take weeks however, so it is worth taking the time to plan your survey so that the data you eventually get is high quality and accurately addresses the research questions you are investigating. This quick guide provides a brief introduction to things you need to consider when designing a survey to help you get the most out of the survey data.
Avoiding survey fatigue among students
Please help us to avoid survey fatigue among students by following our guidelines.
Please remember that students receive numerous requests throughout the year to complete surveys. These include:
- annual surveys organised by the University to help monitor student satisfaction, such as NSS, UKES, PTES, PRES and unit evaluations
- student satisfaction and experience surveys organised by Faculties, the School, Departments, The SU Bath and a range of professional services
- requests to take part in student and academic research.
Following university procedures for running a survey
The University of Bath has procedures in place to manage how staff can publish surveys of students, partly to avoid survey fatigue and also to avoid interfering with central surveys such as the National Student Survey. If students are spammed with too many survey requests, colleagues may end up with a poor response rate. The University of Bath does not support or promote a single individual students survey for their dissertation or thesis, etc, because if the institution did this for one student, we would have to do it for every one of our thousands of students! However, the University does support staff surveying students through specific guidance.
Quick steps to designing and delivering a survey
If you want to survey other students, which would be without central University of Bath support, it is important to consider any ethical considerations first. You must first get ethical clearance to ensure you are complying with GDPR legislation about holding any personal information, asking sensitive questions that might have ethical implications (e.g. if you are asking about crime, there may be sensitivities if people you want to survey may be affected by crime). If you also want to publish anything from your survey you will need ethical clearance to ensure your followed all required procedures, including seeking permission to use participants’ anonymised data.
Planning your survey
A good survey can take weeks (or longer) to develop so allow time for this. The first thing to consider is the length of survey, as creating a survey which is too long with too many questions will put people off completing it. A survey which is too short may not give you all the information you want, so there is a compromise to be considered for the survey design. Some advice on best practice in survey design is available.
Testing your survey
Another important part of survey design is to test it on people first. When you design your survey the questions may make perfect sense to you, but your use of language may not be clear to others so get people to test the survey to make sure your questions are clear and the participant understands how to complete the survey.
Distributing your survey
When publishing your survey you want as many people as possible to complete it so you need to think about how and where you will publish it. You also need to consider the risk of survey fatigue or clashes with other surveys such as the National Student Survey. So you will need to carefully plan how you publish your survey to maximise the response rate.
Analysing the results
Once you have your survey responses you will need to analyse the data. If you have lots of numerical (quantitative) data you may need to use specialist software to help you analyse and report on the data, such as undertaking statistical analysis. Support is available through the MASH centre. You can use Excel for simple calculations but for more sophisticated analysis you may need to use programs like SPSS or R. If you have other forms of data such as text responses (qualitative data), you may need to code the responses and do specialised analysis of the data. For this type of analysis you can use programs like NVIVO and you should seek help on how to code responses and how to use NVIVO for the analysis.
Undertaking a survey may seem challenging but you will learn a lot of valuable new skills in the process. To get help with creating, publishing and analysing a survey you should seek the help of your supervisor initially. The Skills Centre is also a good source of support.
Advice for staff planning to survey students
If you are a member of staff, please see our guidance for staff on surveying students.