UX Research Ethics & Biases

Any user research involves people and their privacy. In contrast to academic research, there are no strict ethical regulations in UX research. However, it's your responsibility to ensure that your research is bias-free, accurate, and respectful of users' privacy and personal data.

Maintaining high ethical standards in your research demonstrates a commitment to high morals and protects your company's reputation. Violating users' confidentiality or acting without their consent can ruin your good name for years.

The digital age has brought us to the point where people can find traces of your ethical transgressions on the internet years after they’ve occurred. This can actively discourage them from using your products.^[2]

Explain the goals of your study and how you will use and present this data to your users. This can make them more comfortable in signing the consent form and get them to share their thoughts more enthusiastically.

Sometimes, you may feel reluctant about mentioning your company's name or the research goals as it may impact how users act or answer questions. There is also the possibility that participants will answer the way they think you want them to.

These factors can influence the findings and impede your research goals. So, if you have any such solid reasons for holding back information before the study, you may do so but make sure you provide the full information to users as soon as possible. This allows them the option of withdrawing their consent.

The ethical approach to user research implies sensitivity to users' feelings. Some users may feel nervous about testing or reluctant to share personal details. This sensitivity is even more essential when your research deals with minorities or vulnerable populations who have experienced trauma, injury, abuse, or general misfortune.

Your goal is to make them feel relaxed and reassure them that their information is safe. If you know that some questions may disclose sensitive data, ensure that only one or two people participate in the session.

Some topics may be uncomfortable to discuss with someone outside of the minority group. Make sure the moderator involved is someone the participants will feel comfortable with.

UX designers are humans, too, and may have biases about users' behaviors, thoughts, and needs. That's why we need user research — to prove our assumptions right or wrong.

When you conduct a research session, avoid asking leading questions to verify your assumptions. Listen to what users actually say or do. Research findings will be insightful only if you observe users instead of trying to reinforce the desired outcome.

When you analyze findings or present them, be honest and accurate about what you've learned during the study even if the data disappoints your team or stakeholders.

The most effective way to communicate your research findings is to use direct quotes. They speak louder than facts and have a stronger effect. However, quotes are only powerful when you support them by the number of participants sharing the same views. Without the numbers backing them, quotes are just opinions of individuals.

An informed consent or research consent form is a documented agreement between a researcher and participant. The researcher provides the full information about the research goals, the scope of involvement of the participant, and how the research findings will be used. The participant agrees to the study terms and makes an informed decision to participate in the research by signing the form.

You might be confused about whether you should ask people to sign the consent form if it’s just a short phone call and can hardly be named a research session. If you're going to use users' quotes publicly or share this information with your team, you definitely should notify users about it and receive their consent.

The consent form also has other benefits:

• It is a consistent way to inform all participants about the research details, especially when they are large in number
• It ensures that the researchers prioritize the welfare of participants and their data
• It motivates researchers to take the study seriously^[3]

Even if you're confident your study won't harm its participants, you should always consider the hidden, indirect harm it can cause, such as manipulating users' emotional states or behavior.

In 2012, Facebook altered the news feeds of 689,000 users for a week, showing them happier or sadder than average content. The study findings revealed that people who saw happier content posted more positive words on their feed. Conversely, users who saw sadder content posted more negative words than usual. While the goal was to improve Facebook services and provide more relevant and engaging content to people, the selected method was unethical. Instead of just observing user data, Facebook manipulated users' feelings like they were lab rats. What’s worse, the company’s data-use policy didn’t mention the word “research” even once when the experiment was conducted. This means users never agreed to participate in Facebook’s studies.^[4]

This example demonstrates how the negligence of research ethics may hurt users' feelings and affect your company's name. For a much smaller company than Facebook, manipulating users' data during a study can have disastrous consequences.

When users agree to participate in user research, they trust you and believe you would keep their data safe and share it on agreed terms only. If you want to mention a user's name even though users prefer to stay anonymous, respect their decision and come up with a pseudonym. Sharing an image of your users or their houses is recommended only if you have permission to do so.

Once the research is finished, consider disposing of user data that is no longer relevant to your project. Consider the legal complications you might encounter if the information is stolen and revealed.

Big data refers to a large volume of digital data transaction processing systems, customer databases, documents, emails, medical records, internet clickstream logs, mobile apps, and social networks. It also includes machine-generated data, such as network and server log files and data from sensors on manufacturing machines, industrial equipment, and internet of things devices.^[5]

Companies collect and use this data in machine learning projects and to identify general trends and correlations. Unfortunately, big data is often used and transferred between companies without notifying users. 79% of Americans report that they were at least somewhat concerned about how companies use the data they collected about them.^[6]

It's easy to think you're not breaking any ethical rules by asking users to agree to your Privacy Policy, but most people don't even read them. According to an Ofcom report, 65% of users accept the terms and conditions and the privacy policy without reading them.^[7] It means people can be affected by big data research, and you will have to face complaints even if you don't legally violate any regulations.

Even if you’re conducting user research without face-to-face communication, it’s essential to remember that big data is still about people. You may accidentally reveal someone's identity by mentioning a combination of their data like birth date, gender, location, and digital activity in your research.

Also, remember that big data can be hard to interpret without context and can lead you to the wrong conclusions. For example, a photo on Instagram can represent an approval/disapproval of a phenomenon, a simple observation, or an attempt to look better in the followers' eyes.

Researchers are humans, and all humans have biases — tendencies, inclinations, or prejudices toward or against something or someone that exists in the subconscious.^[8] When people encounter unfamiliar situations and don't have enough information about them, their brains explain them based on their existing beliefs and assumptions. Biases are our natural ability to process information faster. However, they are often based on stereotypes and can result in snap decisions, bad judgments, and false predictions.

For example, some people often have misconceptions about women's abilities to do technical jobs, even when they have similar skills and job experience as men.

Biases can seriously impact your user research and, eventually, the design and success of your product. One of the most effective methods to prevent biases from both researchers and participants is to ask the right questions. If researchers don't let their assumptions guide participants toward desired conclusions, respondents are more likely to give honest and insightful replies.

Confirmation biases occur when researchers tend to focus on evidence that confirms their assumptions and ignore opinions that contradict those assumptions. Users may complain that microcopy in a product contains too much technical jargon and is difficult to comprehend, for example. Influenced by confirmation biases, UX researchers might discount such feedback because the copy makes sense to them.

UX practitioners can avoid confirmation biases by:

• Asking open-ended questions that inspire participants to talk
• Avoid questions that seek confirmation of your ideas
• Actively listening to what users say
• Including more people in the study to check if their hypothesis is right or wrong

False consensus bias assumes that everyone thinks the same way you do. For example, if you're an Italian and love soccer, you might assume that all Italians love soccer and that those with different opinions are outliers. UX researchers can be influenced by this bias when believing that most users will agree with their idea or design. It is a dangerous bias that might result in huge expenses and wasted time.

You can reduce the risks of false consensus bias by exploring your users' needs, pain points, and preferences. In the early stages of product development, you should also define and validate your assumptions with real or potential users. It can prevent your team from building a product no one actually needs.

Our brains naturally remember recent experiences better. In UX research, recency bias occurs when a researcher gives more weight to what participants say or do by the end of the interview or at the last usability testing session. As a result, the research findings may miss insights that occur at the beginning of the session.

You can avoid this bias by taking notes or recording the session with the participants’ consent and reviewing it later.

Our brains tend to give more weight to people we meet first in a new, unfamiliar situation. In a UX study, primacy bias happens when UX researchers remember the first respondent more strongly than others.

Taking notes and recording your session with a participant’s permission can help you overcome this bias by interpreting the research findings more thoroughly and treating all participants’ answers equally.

Implicit bias is an attitude or stereotype we have subconsciously about certain groups of people, usually of a different race, gender, social status, education level, or sexual orientation.

Like all people, UX researchers also have preconceived notions and generalizations and may treat people in diverse groups differently. For example, UX practitioners can talk more slowly or louder to older people during the study, assuming they all have hearing or cognitive disabilities. This can make your participants feel discriminated against and impede them from being honest when answering your questions.^[9]

Before the session, list all your assumptions about this certain group of participants and skeptically reflect on them. Your main goal should not be to prove your ideas but to find what is actually going on inside users’ minds.

The sunk cost fallacy describes the tendency to keep investing in our original choice because we've already spent a lot of time, effort, or money on it. Even if its costs outweigh its benefits, we feel guilty and reluctant to give up.^[10]

It often happens to designers who get emotionally connected to their designs. They find it impossible to admit that their solution doesn't solve the problem and must be changed. When UX researchers get affected by this bias, they continue conducting research, collecting data, and obsessing over the findings, even if they've strayed off the course.

To avoid this bias, define your research goals beforehand and outline points at which you can evaluate whether you're on the right track.