Bias can skew results and lead to flawed strategies (argh!)

In the ever-evolving world of marketing, research is often the cornerstone of a successful marketing strategy. Done well, research can explore consumer behavior, market trends, brand perception and new concepts and provide us with insights that elevate our offer. But if we’re not careful, bias can sneak its way into our research skewing results and ultimately leading to flawed marketing strategies and goals.

Here are a few more examples of how bias might creep into your market research:

• Confirmation bias: This occurs when you subconsciously favour information that confirms your existing beliefs. For instance, your company is heavily invested in a specific product, and so you focus in on the data that supports the product in your report.
• Wording bias: The way you phrase questions and answers options can significantly impact responses. For example, asking "Are you in favour of a new hospital that will improve services and create jobs" will lead to different results than asking “Are you in favour of a new hospital?" because the first option includes leading language.
• Recency bias: People have a bias towards more recent information. If you ask a participant about their experience of something over the last year, they will actually tell you about what they experienced very recently.
• Social desirability bias: This occurs when respondents are asked more sensitive questions and answer with what they think is a socially desirable response, rather than with honesty. For example, if you are doing a healthy living survey and you ask “How many units of alcohol do you drink a week?” participants are likely to give a low figure, regardless of the truth, because they are confirming to what they think is more socially desirable in this context – drinking less.

Five ways to minimise bias (and get more reliable results)

So, how do we mitigate bias and gather data that is as reliable as possible? Here are some things to think about:

• Consider how you recruit your participants: Is an email to your customer list going to provide a balanced set of responses? Or do you need to combine this with an ad or telephone recruitment campaign that will reach non-customers? Aim to reach as wide an audience as possible within your target market. That way you get a more representative sample with diverse opinions.
• Be careful with question order: Sequence your questions thoughtfully to avoid early questions influencing the responses to later ones. And consider randomising the order of multiple choice and matrix questions.
• Use neutral language: Avoid leading language or emotional triggers when you ask your questions.
• Include an open-ended question: Having at least one free text question gives participants the opportunity to put a response in their own words rather than yours.
• Be open with your findings: After data collection, critically analyse your findings and consider potential biases that might have crept in. Provide a complete view of your results and make sure you don’t gloss over anything that goes against what you were hoping to prove. Better to know now!

Avoiding bias in market research is a challenge, but by being mindful of its existence and doing your best to mitigate it, you can produce research that delivers valuable, reliable insights. And this in turn empowers you to make more informed marketing decisions, that resonate with your target audience and propel your brand forward.

By: Rosie Pratt