They’re the stuff of movies, TV shows, and usability labs.
One-way mirrors (or two-way mirrors depending on who you ask) are an enduring symbol of interrogation, psychology experiments, focus groups, and usability tests.
This special piece of glass is brightly lit from one side to allow people to inconspicuously observe people on the other side.
The technology is simple and actually quite old with a “transparent mirror” patented in 1903.
Researchers once considered it essential, but in the last few years I’ve seen companies moving away from the mirror. Here are some advantages and disadvantages of the one-way mirror.
- Visibility: With the right viewing angle, you can see the entire room in context, and thus capture all meaningful information such as the emotional expressions of the participant and interactions with the moderator. Visibility is especially valuable when you’re observing participants interact with physical products (like setting up a TV) and move around the room.
- Low tech: A one-way mirror doesn’t take much maintenance after it’s set up (no power, cables, or converters needed).
- Looks cool: There’s a reason one-way mirrors make appearances in movies and TV shows—they look cool! They were one of the highlights of lab tours when I worked at Oracle. Having a one-way mirror makes it at least “look” like you’re doing serious research.
- Better than a room full of people: Looking through a one-way mirror is better than having a bunch of people crammed in the same room or looking over a participant’s shoulder. It’s also less distracting for participants because observers aren’t moving around and talking.
- Intimidates some participants: One-way mirrors are hardly subtle, and participants generally know their purpose (participants watch TV and movies too!). To some they are perceived as “scary, intimidating, nerve racking, and uncomfortable.” This perception may affect participant behavior, although more data is needed to understand what effects, if any, there are.
- Limits the view: It can be hard to see participants’ computer screens from behind the mirror. What happens on the screen is usually of most interest to observers in usability sessions. Consequently, observers end up looking at the feed of the participant’s monitor. Which can be sent to anywhere in the world. In fact, a lot of what you can get from a one-way mirror can be replicated with well-placed cameras and direct feeds of computer screens.
- Requires a large dedicated space: For the one-way mirror to work, you need the observation room to be adjacent to the lab. This usually requires a lot of dedicated space, which for many urban areas comes at a significant cost.
- Increases noise and light pollution: Being next to the participants to view through the mirror means observers have to keep their voices down (no laughing or loud talking) and keep the room dark. Not being in the room with the participant reduces distractions, but not completely. Open doors, computers, and phones generate light that allows the participant to see the observers, which can create awkward distractions for the participant.
Does the Mirror Affect Behavior?
Each research team needs to weigh the practical advantages and disadvantages to a one-way mirror. But one question to consider is to what extent a one-way mirror actually affects behavior and how much it may affect your research conclusions.
When we looked in the literature we found little discussion on how one-way mirrors affect behavior. We did find some research on the effect of mirrors themselves. After all, when participants look in the mirror, they see themselves looking right back, which has consequences. We found that mirrors increase self-awareness, which leads to
- Less cheating on a puzzle with a mirror; kids were also more altruistic.
- An increase in the reports of negative emotions [pdf].
How a one-way mirror may affect the data from usability tests, in-depth interviews, or focus groups is still an open research question. Anecdotally, I haven’t noticed a difference in behavior from the participants with or without the one-way mirror. Jim Lewis also echoed his experience of it being a negligible influence but did offer this memorable example
Out of several hundred people that I observed over those years, I only had one who was so nervous that we wound up sending her home. There was one guy who participated in a printer study that involved loading fanfold paper forms (remember that?) and aligning them so the information would print in the right areas. He got so frustrated with the task that at one point he yanked the paper out of the printer, wadded it up, and threw it at the one-way mirror (on that occasion, there was no facilitator in the room). That’s the only time I can think of where someone appeared to be aware of the observers during task performance—most of the time they just appeared to be completely engrossed in the task.
One-Way Mirror, Mirror on the Wall…
In our lab at MeasuringU we have a full digital setup and decided against the one-way mirror (the cons outweighed the pros for us). But if you have a one-way mirror (or decide to install one), here are some best practices based on other researchers’ experiences.
- Rather than trying to pretend the mirror doesn’t exist, point it out and explain its use, and provide the chance to address participant concerns.
- Consider moderating the sessions from within the lab as opposed to being the voice of god—although more research is needed to understand the effects, if any, of the location of the facilitator.
- Have participants sit somewhere where they are not directly facing the mirror. The less opportunities participants have to look at it, the less they will be reminded they are being observed.