The Hawthorne effect and its impact on UX research
Have you ever wondered whether simply observing someone can change their behavior?
In user experience research, this isn’t just a theoretical question, it’s a real situation that can directly affect what we discover during a study.
This phenomenon, known as the Hawthorne Effect, dates back to the 1920s, when researchers were studying how lighting affected productivity in a factory. Interestingly, workers improved their performance even when lighting conditions worsened. In the end, what made the difference wasn’t the light itself, but the fact that they knew they were being observed.
Years later, sociologist Henry A. Landsberger revisited these findings and coined the term Hawthorne Effect, associating it with the special attention participants received. While the original studies have been questioned in terms of validity, the core idea remains relevant: being observed changes behavior.
And in UX, where we closely observe people as they interact with products or services, this is something we cannot ignore.
How does the Hawthorne Effect show up in UX?
The way this effect manifests varies depending on the type of study. Whether we’re doing usability testing, interviews, surveys, diary studies, or observational research, one thing remains constant: people don’t behave the same way when they know someone is watching.
They may try to “do the right thing”, act more patiently, avoid mistakes, or give answers that sound better than they actually are. Since we can’t eliminate this influence entirely, what we do is design our research to reduce it and interpret it in context.
Here’s how we address it across different types of UX studies:
Field studies: when observation can alter the everyday
In field studies, we aim to understand how people behave in their natural environments. But if they feel they’re being evaluated, they may hide shortcuts, mistakes, or personal ways of completing tasks. After all, who wants to admit they improvised a “non-standard” but more effective solution?
What we do to avoid distorted observations:
- We’re careful with our tone: The key is to observe without judgment. No corrections, no comments, just respect.
- We extend the study duration or conduct multiple sessions. This way, the researcher’s presence becomes part of the environment.
- We build closeness without intrusion. Gaining trust takes time, but it’s essential for authentic behavior to emerge.
User testing: more “polite” behavior than usual
During a test, it’s common for participants to be more patient or persistent than they normally would be. Even when something doesn’t work, they may try to push through so they don’t “look bad.” Although we clarify that it’s the design being tested, not them, our presence still influences their behavior.
How we minimize this effect:
- We design tasks that match real-life situations, keeping them as natural as possible.
- We repeat multiple times that we’re not evaluating the person, but the product.
- We maintain a friendly, relaxed tone, avoiding excessive politeness that might pressure them to “perform well.”
- We value every comment, especially the critical ones, as those are often the most useful.
Surveys: answers that sound good, but aren’t always true
In surveys, it’s common for people to respond with what they think they should say, rather than what they actually do. This is especially noticeable on sensitive topics like health, privacy, or consumption habits.
How we handle this:
- We combine direct and indirect questions to detect real attitudes without forcing explicit responses.
- Whenever possible, we ask for concrete examples, images, or descriptions to complement answers.
- We explain why we’re asking each question and how we’ll use the data. Knowing the purpose helps participants respond more confidently.
Diary studies: narratives that filter reality
Diary studies are incredibly valuable, but participants may unintentionally edit their experiences. They might forget to record key events or only share what seems “acceptable.”
Our approach in these studies:
- We’re present but not intrusive. We want participants to feel supported, not monitored.
- We simplify data collection: fewer forms, more focus on essentials.
- We adapt the format to digital familiarity, making it easy to use and barrier-free.
- We show examples of how their input has influenced real decisions, which builds trust and encourages honest sharing.
So… is research still worth it if we know behavior changes?
Absolutely, no question. Observing people, even with some bias involved, is always more reliable than making assumptions.
The key lies in being aware of those nuances and designing our research with them in mind. Good research isn’t just about applying a method, it’s about creating a space where people feel safe to be themselves.
Observing without judgment, designing with empathy, and carefully shaping every part of the process is what truly brings us closer to understanding how people experience a product. And that understanding, with all its imperfections, is where the real value lies.
This is a translation of the following article from our corporate website:
