The process of designing the survey really helped to narrow down my research. Initially coming up with questions made me want to go much deeper and wider with my inquiry but I had to make sure I stayed within the scope of this project. So part of this process was deciding where to set my focus and make sure the data I (hopefully) receive in return is useful and meaningful.
The main areas I am investigating:
– Level of comfort talking about menstruation-related issues in different environments
– Perception of stigmas and taboos surrounding menstruation
– Impact menstruation has on study and work environment
The overall structure
Section 1: General data (age, gender etc)
Section 2: To be filled out by persons who menstruate
Section 3: To be filled out by persons who don’t menstruate
Section 4: Indicate interest for follow-up interview and submit
In order to get useful data from the questions I needed to find the right measurement to quantify the answers, so that the answers are comparable between the questions and different sections of the survey. I played with several ways of answering the same question as well as changing the questions themselves in order to get more precise results.
Example question:
Do you feel comfortable talking about menstruation in a private environment?
No
Yes – sometimes
Yes – most times
Yes – always
After discussing the survey structure with my partner it became clear that terms like ‘sometimes’ and ‘most times’ are very subjective. In addition I felt that its not so much about the amount of times one is comfortable talking about menstruation but the degree of comfort with the subject. Therefore I changed the question to:
How comfortable do you feel talking about menstruation in private?
I considered a 1-5 scale (Birkett, 2021) but I decided that I didn’t want the answer to be a number (e.g. the average comfort level is 2.6) but something more relatable that gives words to a subject that is not often spoken about. After more research on surveys and looking into how to ‘measure’ comfort by looking at mental health surveys online I adapted a sort of 1-5 scale of words that felt more in line with the data I wanted to get from the question:
Very uncomfortable
Fairly uncomfortable
Fine
Fairly comfortable
Very comfortable
Prefer not to say
[Edit (5th Nov): Now looking at it again after having launched the survey maybe ‘Fine’ wasn’t the best medium answer and something along the lines of ‘Neither comfortable nor uncomfortable’ would have been even better but I still think it works for what I am trying to achieve.]
Yet the term ‘in private’ was still not very clear, especially as this is meant as a control question to the following one that asks about ‘at university’.
So for both of them I added clarification of the type of setting the question refers to.
Finished questions:
How comfortable do you feel talking about menstruation in private (e.g. at a friend’s house)?
Very uncomfortable
Fairly uncomfortable
Fine
Fairly comfortable
Very comfortable
Prefer not to say
Making it inclusive
I initially struggled a bit with consistency of language, how to make it inclusive but also be precise and keep questions short and easy to understand.
A lot of my questions had the words ‘male’ and ‘female’ in them for ease of reading and I had a disclaimer at the top of the survey that clarified “’female’ or ‘women’ meaning any menstruating person (currently or in the past), including those who identify differently.”.
I had listened to an episode of ‘Gyncast’ (Der Tagesspiegel, 2020) about menstruation and attended the UAL ‘Menopause Matters – Lunch and Learn” event (19th Oct 2021) which had similar disclaimers in the beginning, however my partner pointed out that making the language of this (written) document inclusive really wasn’t that difficult seeing that I mostly used inclusive language already. So in my last overhaul I made sure the language throughout is inclusive of all persons who menstruate and I deleted the disclaimer.
Outside of the overall language I spent some time looking into how to ask about gender in a survey. Gender is highly relevant to my research and I wanted to make sure I asked the question inclusively. As for the question itself I steered away from asking ‘What is your gender?’ and instead asked ‘How would you describe your gender?’.
I found lists of up to 25 different gender identities (Netigate, 2021) but eventually settled on a list of three answers and a ‘prefer not to say’ option.
One of the options was going to be ‘Prefer to self-describe as…’ with an open text field, however when I came to creating the survey in Microsoft Forms I found that it doesn’t allow a mix of multiple choice and text field answer. Also keeping in mind that I want the data in a format that will be easy to analyse I settled on the following list:
Female (including transgender women)
Male (including transgender men)
Other / Not listed
Prefer not to say
Short & Sweet
Throughout designing the survey I kept analysing what kind of survey I would participate in myself.
Probably one that doesn’t take too much time and for which I know in advance roughly how much time it will take me to do. Also one that potentially ‘gives’ me something in return like access to the research findings or an outlook to how the research might influence my life going forwards.
During the second PgCert workshop I was able to test the survey with some of the PgCerters. I received 8 responses with an average time to complete of 7min43.
Assuming that some students may take longer to read the questions, especially those whose first language is not English, I assumed an average time of 10mins. This worked well with my thinking that 10mins felt doable if someone asked me to volunteer my time for their research, but much longer might put me off.
I made sure to mention that it would roughly take 10mins to complete in the short paragraph introducing the survey to students and staff.
Summary
Overall I am happy with the way I designed the surveys. Because I was going to the UAE for a project for all of November I rushed getting them out a little bit though and launched the student survey before finishing the questions for the staff survey. Holding off on that would have helped finesse the comparability some more.
They are still very similar but there are a few adjustments I would have made in the language. For example in the student survey I refer to ‘peers’, in the same question on the staff survey I refer to ‘colleagues’ (questions 20 and 26 respectively). I will still get meaningful data but just have to keep this in mind for the analysis.
References
Alchemer (2021). How to Write Gender Questions for a Survey. [Blog] Alchemer Blog. May 20, 2021. Available at: https://www.alchemer.com/resources/blog/how-to-write-survey-gender-questions/ [Accessed 20 October 2021]
Birkett, A. (2021). Survey Response Scales: How to Choose the Right One. [Blog] All Things Data-Driven Marketing. Last updated: Mar 15, 2021. Available at: https://cxl.com/blog/survey-response-scales/ [Accessed 18 October 2021]
Lewis, H (2018). Avoid the awkward: asking about gender. [Blog] Thrive Blog. 5th January 2018. Available at: https://wewillthrive.co.uk/resources/blogs/avoid-the-awkward-asking-about-gender/ [Accessed 20 October 2021]
Netigate (2021). How to write a good survey introduction for email. [Blog] Netigate. Last updated: 16 April, 2021. Available at: https://www.netigate.net/articles/survey-tips/how-to-write-better-email-texts-for-your-survey/ [Accessed: 22 October 2021]
Der Tagesspiegel (2020). ‘Wie funktioniert der Zyklus?’, Gyncast. [podcast] Available at: https://podcasts.apple.com/us/podcast/wie-funktioniert-der-zyklus/id1509255500?i=1000474512211/ [Accessed 15 Oct. 2021]