How to Avoid Bias in Market Research
We like what we like. That’s the way I’m wired—that’s the way we’re all wired. But when it comes to market research or validating the next big idea, it’s important to remove our bias as much as possible so the truth can come to the surface. This isn’t easy but will help you become a better researcher or innovator.
The trick to avoiding bias is asking questions in such a way that allows your consumer to think beyond their bias without imposing your own bias. Whether you are doing consumer interviews or a focus group, you need to address bias upfront. Let’s look at the 5 types of bias and how you can avoid these…
Bias in Research
The problem with imposing your own bias in research is that it tricks the other person (the consumer in this case) into answering as if they agree with you, when in fact, they don’t. But the opposite end of the spectrum leaves the door open for consumer to answer in such a way that embraces their own bias.
You always want to ask engaging questions and avoid bias in business writing, but even more so when designing research questions. Questions that present unintentional bias will skew the answers you receive and give you results that don’t correctly represent your consumer base. There are ways to avoid bias and we’ll talk about those below, but let’s first talk about the different types of bias.
The Different Types of Bias
There are many types of bias that can happen. Here are a few short descriptions:
Demand Bias: This generally happens when the consumer eagerly provides the ‘right’ answer just to make you happy. While these might be what you want to hear, they’re not truly reflective of how the consumer actually feels about a topic. You may have already experienced this when testing new ideas and concepts. Check out this example. “Our team spent the past 2 years designing this new product to solve X. What do you think about it?” This doesn’t leave much room for consumers to truthfully respond. No one wants to squash your dreams and 2 years’ worth of work.
Social Desirability Bias: This is when people feel pressured to respond in a certain way because they feel it’s socially acceptable or desirable, as in the above example. This can also happen when you impose your own bias in framing the research experience, the questions you ask, or even in the methods you choose. For example, you may ask a question framed with a social (or in this case, environmental) issue that the consumer may not be personally motivated around. “Oceans are increasingly being filled with plastics and killing wildlife. What do you think about sustainability in packaging?”
Dissent Bias & Acquiescence Bias: Dissent bias is when someone answers every question negatively, and acquiescence bias is when someone answers every question positively.
Extreme Responding & Neutral Responding: Just like dissent bias and acquiescence bias, extreme responding and neutral responding is when your consumer solely responds on either extreme of a Likert Scale question or answers straight down the neutral answers.
Personal Bias: Again, everyone has a personal bias—we are all biased in thinking that we’re better than we are.
5 Tips to Break out of your Bias box
To avoid bringing bias into your research, look to methods that allow your audience to respond while in their natural environments and without the influence of people around them, even the moderator. Vurvey helps enable this by creating an asynchronous responding experience. Consumers can respond to questions while at home, on their own devices (i.e. smartphones, laptops), and share using audio and video. This creates an intimate and safe environment for consumers to share how they truly feel without judgment.
Now that you have the method selected, let’s look at 5 tips to avoid bias in framing and asking questions.
With yourself and about your actual customer base. Nothing can sink the research ship faster than researching in the wrong place or with a narrow subset of people. If you’re in the discovery phase of research, keep your responding pool open to hearing insights beyond who typically purchases your product or service. You might be surprised by what you hear.
Let’s look at men’s underwear, a $30B global market opportunity. Want to take a guess at one of the largest cohorts buying men’s underwear? Women. Whether for their spouse, partner, or kids, women are searching for men’s underwear, evaluating claims, testing materials, and even making the final purchasing decision.
Removing biases around who your audience is can help include consumers who are a part of your actual customer experience, even though they might not be the end-user of your product or service. Customer relationship management tools (CRMs) can help with sending your Vurvey to the right target audience and allow you to divide your consumer base into cohorts. For example, you may have a different set of questions for the buyer versus the end-users. Always ask the question, who might you be excluding?
If your research or feedback loops are branded, changing the overall tone of your copy or questions into something different than your ‘brand speak’ could unintentionally bias the experience for customers.
For example, if your company’s tone is fun and upbeat, but the questions sound dry, disengaged, and corporate, you might have a difficult time getting candid answers from your consumers. They may feel like they’re talking to “the man” behind the corporate wall—even if there isn’t one.
Being yourself (or your brand) helps establish trust and that their feedback is valuable. The watch out here is going too far in ‘branding’ questions to avoid overly biased and positive responses.
Less is more. Break compound questions down into a set of simple, easy-to-understand sentences. It’s vital to the success of your research that questions are not complex or leading. Unnecessarily complex questions invite unnecessarily complex answers.
Vurvey’s Multiple Choice + Video question type helps keep this process simple. Consumers can provide a rating to the experience, and then use video to talk through the experience and all of the details. Asking Why? after the initial question helps consumers keep sharing and talking. This is always a great way to keep your market research concise, but also open to new discoveries. Conversely, concise questions lead to direct answers. Keep your questions short and to the point so your consumer knows how to answer you in a succinct way.
Have you recently shopped at Target?
What does “recently” mean in this question? If there’s debate about whether “next Wednesday” means the literal next Wednesday on the calendar or the Wednesday of next week, then you can bet the words “recently” and “shopped” may trip people up. Vaguely worded questions will take the quality out of your desired responses.
As you’re designing questions, make sure your language is as precise as possible. Eliminate terms that might be confusing or open to interpretation. This question might work better as “Have you purchased anything from the Target app in the past 60 days?” This clarifies that you want to know specifically about purchases vs browsing, that the activity took place in the app vs retail store, and that it happened in the past 60 days.
Of your structure, that is. Pay attention to the structure and order of your questions. At the micro view, some questions can build on each other to create a “full picture” aspect for you. At the macro view, you’ll want to organize your questions so the thoughts themselves build on one another, leaving the deeper, loaded, and open-ended questions for the tail end of your Vurvey. This allows your consumer to build to those moments so they can be mentally ready for truly candid and vulnerable answers.
Test and iterate!
Once you create the questions in your Vurvey, the most important part is to Preview the experience.
You might preview this on your own, share it with team members, or even send it out to a few consumers as a test run. ‘Playing it out” like this helps gather vital feedback to see if your questions worked as intended. Did your structure, wording, and prompts help consumers respond to the specific areas that you were interested in learning about? If they didn’t, tweak the questions and test them again.
Thanks for reading this post. If your planning your next research project, try asynchronous platforms like Vurvey and these 5 tips to help avoid bias in your experience. Happy researching!