The perfect UX research recipe 🍳
** Mixed Methods Research
Hello Human
Your favourite researchers are back from the forced hiatus. For that don’t know, we decided to go on a break to focus on what was happening in our country (the #EndSars movement in Nigeria).
So before we start talking about recipes and mixing methods, we want you to pause. Take a deep breath, exhale and genuinely answer this question: How are you doing?
We genuinely want to know how you all are doing so if you’re comfortable doing so, you can send us your response or tell someone you trust or just tweet it.
Your mental health is important to us, so please be intentional about evaluating it from time to time. 2020 has been one heck of a year, but we hope you are doing reasonably okay and staying safe. We are too.
What the heck is ‘mixed methods’? Okay so mixed methods. I’ve had a lot of people reach out to me to ask me which particular research methodology they should use for their projects. When people ask this question, they hope to get one particular research methodology they can use. While valid, the question is somewhat myopic and I will tell you why in a minute.
In our last issue (https://mailchi.mp/f96c20d582ba/so-what-forms-of-research-can-i-do) , we talked about research types. We talked about the 4 main ways research is categorized: behavioral & attitudinal, primary & secondary, foundational & evaluative, and qualitative & quantitative. Qualitative & Quantitative categorization is most common of them all, and mixed methods capitalizes on them both. The idea behind mixed methods is using both qualitative methods and quantitative methods to answer a set of research questions.
Huh? What does that even mean?
Let me explain. I’m sure you’ve heard people say this, but I’m going to repeat it: Quantitative methods answer the “what” questions why qualitative methods answers the “why” questions. For some reason, people like to treat them like they’re mutually exclusive. So they’ll say things like, “I prefer quantitative methods because it gives factual responses, qualitative methods is too fairytale-like and unreliable.” What a lie. I’ll give you an example:
For a given product, data from Google Analytics told us that the product has a 23% conversion rate, with 45% of users dropping off after clicking the “Checkout” button.
This insight is quantitative because it has numbers attached to it. This insight has told us what users do, but we’re still not sure why they do it. Why do 45% of users drop off after clicking the “Checkout” button? Why does the product have a 23% conversion rate? These are questions only qualitative research can answer. Assuming some user interviews (qualitative method) were conducted to understand why the drop-off number at the checkout point is so high, and the research revealed that:
Users struggled to checkout their items because there were some glitches on the checkout page that confused and frustrated them.
Now we have the full story. We can even combine the qualitative and quantitative insight to say:
The product has a conversation rate of 23%, and the checkout page has a drop-off rate of 45%, which can be attributed to the glitches in the checkout page that ends up confusing and frustrating users.
Do you see the difference? Because we chose to combine qualitative and quantitative methods (or mix methods), we were able to get a holistic understanding of the problem and what some plausible solutions could be. The way I see it, using either qualitative or quantitative methods only gives you 50% of the story. For you to get 100% of the story, you need to mix methods.
What does this mean for me as a product person? In the example mentioned above, if the designer had ran with the quantitative insight only, s/he would have made changes to the design of the checkout page based on guesswork & trial and error. On the other hand, if s/he had ran with the qualitative insight only, s/he would not have been able to quantify the problem, and measure the impact of the problem and plausible solutions on the business.
A disclaimer though, mixed methods isn’t one-size-fits-all. Not all research questions require mixed methods. If your research question is around “what button to use” or “what colour to use” or “where to place this button”, mixed methods wouldn’t deliver as much value. Questions like this can be easily answered with a single method (commonly quantitative).
Mixed methods is particularly useful in answering complex & strategic questions. I find mixed-methods very valuable during discovery research because of the number of ambiguities in discovery research, and the strategic nature of the insights mixed-methods proffers. I love talking about mixed methods because people are so oblivious to it.
A lot of designers don’t know that it is possible and even beneficial to use more than one research method at a time in a single project. Hearing it makes me really sad. There is no limit to what you can do and how you can mix research methods so by all means, learn and explore.
Closing Thoughts? Shalaye aside, it is evident that mixing methods has a lot of advantages. Knowing all these fantastic things about mixed methods, why don’t people use it more? A major reason why is because they think it is stressful, expensive and adds an extra layer of complexity to an already complex discipline (the discipline of user research).
A lot of designers and product professionals find user research very dull and stressful, and I don’t even blame them. To be honest, research stresses me out every single day, and I’m a full-time Researcher so believe me when I say I get it. The thing though is that mixing method gives you value for your money. The insights from mixing methods is boundless and super valuable, so use that as your motivation.
Contrary to popular belief, mixing methods is not expensive. With proper planning, you’ll be able to do cheaply and with ease, and that’s the truth. We’re not telling you to conduct 20 interviews and review millions of quantified data points instead; you can do 5-7 interviews and review thousands of quantified data points.
Thinking about things this way allows you have the best of both worlds as opposed to focusing on getting qualitative insights or quantitative insights only.
In the coming weeks, we’ll be diving deeper into the qualitative and quantitative methods.
Next week, we’re going to be learning about user interviews and how to interview like a pro. In the meantime, check out this article (https://dscout.com/people-nerds/mixed-methods-research) by DScout’s Ben Wiedmaier on mixed methods.
If you learnt something new, don’t forget to share!
Till next week, kids! Your Favourite User Researchers (Lade & Dumss)
Rest In Peace Oke Obi-Enadhuze, you live on in our hearts