UX Tea Break: How to create bulletproof survey questions

[@gamma_girl]

I'm asked to do a survey and this came at a perfect time. Thanks for suggesting cognitive interview. The response "ya I know the time" got me 😃

[@DavidTravis]

Thanks for sharing.

[@zak1993]

This guys a master of UX lol he deserves much more views

[@sayyidadnanhashmi3614]

These UX tea breaks are awesome

[@DavidTravis]

Thank you!

[@anafaria5281]

Hi David, the cognitive interview seems valuable for testing questions to ask participants in user interviews as well. I have two questions that I would like you to answer in the UX Tea Break, if possible: What are the best practices to conduct user interviews to create personas? Also, can you give some examples of questions? Thanks for your attention.

[@DavidTravis]

Thanks for these.

[@payelbandyopadhyay2866]

I really liked what you presented here. Can you do a video on how to calculate sample size for surveys and when to use them? For example, how can I use survey in early stages of product as well as later.

[@DavidTravis]

Here you go: www.surveymonkey.co.uk/mp/sample-size-calculator/

[@shreedharkasbekar9929]

Dear David sir, Can you make a video for ux budget and resources planning?

[@clairemhle1]

Amazing videos! Thank you so much! Just one request - could you kindly change the voice recording device? Your voice is soothing but the machines captures it badly and it sounds like super high pitch. Thank you always for your great videos.

[@DavidTravis]

I'll try my best.

[@samssalman]

Very useful thank you. What I often see is people exclusively using tools like usertesting and adding some questions to go beyond the " feedback on a prototype usability ". Systematically, I see just one follow-up question; "why did you answer this way?". And often the users'reply brings not much additional value or deepens the initial answer. I understand from your demonstration that this is not enough. Thanks again

[@DavidTravis]

You're welcome. 😊

[@samssalman]

@@DavidTravis the idea is to test the surveys if my understanding is correct. Would you recommend following the same practices used for usability tests; testing the survey with 5 users and iterate before considering the survey to be viable?

[@LindaUgbo]

Asking question like how much do people really know about this kind of service, products, or even this type of ? And what has made people use this ? Why is this not the best already ?. Who did this help of favour. How is this protected from sabotage and which area did vulnerable occur ? Think is a good way to emphatise the need at end of the UXUI journey from the beginning of the design.

[@user-lj4lo7cx7m]

Amazing content. Thanks for sharing it!

[@DavidTravis]

Much appreciated, thanks for the comment. 😁

[@geoffwilsoncomedy]

Another brilliant and easy to digest video. However, I noticed that you didn't expand very much on the third question about its purpose, an example, and the desired outcome / takeaway from their answer...

[@KasumiL5x]

Very informative and interesting. Thank you!

[@DavidTravis]

Glad it was helpful!

[@cupofmqtcha]

Thank you so much! Very useful.

[@DavidTravis]

Glad it was helpful.

[@leonardocarvalhoalexandre1688]

tks for sharing such a great video David! best regards from a brazilian UX

[@DavidTravis]

My pleasure! 🙏

[@tarmac5482]

What I've known about Cognitive Interview is that it is for re-creation of events and find possible correlations or misrelations in an occurrence of an event by its participants. I am finding it difficult to relate it to UX? I would really like if you can expand on it. Thank you for sharing though :)

[@DavidTravis]

Maybe you're getting it mixed up with another method? It's been standard practice in survey creation for many years. Here's an article I wrote on it: www.userfocus.co.uk/articles/cognitive-interviewing-for-surveys.html

[@dbsvirtualdesign3374]

Is it necessary to copy a survey questionnaire in an adapted study, or can you just base your questionnaires on how they come up with and just use it as bases for your questions ? i need reply sir. pls pls pls thank you in advance

[@DavidTravis]

I'm not sure I've properly understood your question, but are you asking if you can re-use surveys? In some cases, this makes sense: for example, if you have a standardised survey or a survey that you've used to track user satisfaction over time (like the SUS), then it makes sense to reuse the same survey so you can benchmark improvements. But there are other situations where you can't just re-use what's gone before and need to ask new questions.

[@Littl3BitStrang3]

Really useful and supportive as always :)

[@DavidTravis]

Glad you think so.

[@shoshor377]

Does this mean I follow up with each survey participant with an interview?

[@DavidTravis]

No. You should pre-test your survey with a small number of the kind of people you expect to take it. You fix the problems you find and then you send out the survey to your respondents.

[@ddewjr]

The problem is not with the survey question, but with the response options David suggests. Always collect interval or ratio level survey responses -- How many times have you talked to your physician in the last 6 months ____? Rather than asking for categories of responses like 1 to 3, 4 to 6, etc. That is is a basic rule of data collection. I don't know why he is suggesting such a slipshod way of collecting data. My undergraduate students are taught better than what he accepts in this video.

[@DavidTravis]

I respectfully suggest that you've completely misunderstood what I'm talking about in this video. I'm describing a technique for understanding how users understand our questions, so we can improve upon them. Changing the response option to a free form answer isn't going to help if people don't understand the question in the first place.