Week 27: 18th – 24th March
I’ve been considering the possibilities of interviewing artists about their practice as part of my research methods. This has opened up a whole new element of ‘social research’, which is a way of working that I’m not yet familiar with. Given that this would mean yet more research into designing and implementing these new methods, I was unsure whether to proceed with this plan. However, after working on the Artists Book Fair, my supervisor asked me to help design a feedback questionnaire to help with evaluating and planning, so I realised that this would be a useful addition to my skillset in general.
I’d previously noticed the use of online surveys shared through social media so I decided to look into these as an option by asking my network which were best value according to price vs functionality. Top of the list were Survey Monkey and AYTM, and asking on Twitter even alerted other providers to contact me from eSurveyCreator and Fluid Surveys.
All of these looked interesting and offered different solutions, however in the meantime, my supervisor had been making enquiries of his own and found that the university had a subscription to Bristol Online Surveys (BOS) that we could use free of charge, so we decided on that option. As I have no previous experience in creating surveys, I decided to create a summary of the pdf provided by BOS about designing surveys to get the best results:
Keep surveys as short as possible – the shorter the survey, the more likely you are to hold the respondent’s attention and hence have the survey completed. When creating your survey, consider the following:
* What is the purpose of the survey?
* What kinds of questions does the survey need to ask?
* What sorts of actions are being considered based on the results of the survey?
* Who is the intended audience?
* Are the questions relevant to the respondents?
There are six different types of questions that can be created in a BOS survey: Selection list, multiple choice, multiple answer, single line (free text), multiple line (free text) and date selection. With regard to these question types, take note of the following issues when designing your survey:
* Should a question be mandatory or optional? If you wish to make a question mandatory, then be sure to provide options such as “Not Applicable” or “Don’t know” in case a respondent cannot answer the question.
* Try to avoid an odd number of possible responses to a question as respondents tend to pick the middle one.
* Ensure that question responses use consistent wording.
* Ensure consistency of case (i.e. small or capital letters) in question responses.
All surveys should have the minimum of three pages; containing an opening or welcome page, at least one page which contains questions and then a final thank you page with no questions on it. The welcome page should contain approximate time taken to complete the survey (and the number of pages), an introduction to the survey and basic instructions to help the respondent to navigate through the survey.
Using your survey results
When launching the survey, enter the highest number of respondents that you expect. This number is used to calculate a response rate when viewing survey results and can be modified after a survey has been launched. All results are totalled within each question by scrolling down the page, but there is also the option to:
* Cross reference questions to see the correlation of their answers
* Export results in CSV format
* Filter results by answers to specific questions
* Compare surveys
* Step through individual responses
I’ve yet to launch my first survey but I’m hoping this info helps me to get the results I need.