The observer-expectancy effect

Or, Considering the realities of context

In our studio class, my teammates (Lisa and Kate) and I are currently looking into ways in which the waiting time at Pittsburgh bus stops can be made a less frustrating experience.

In her post, Lisa talks about our strategy for initial exploratory research. Initially, we approached people together, introduced ourselves are students working on a school project, and asked if they had a few minutes to answer some of our questions. It quickly became clear we weren’t getting rich enough information from these people. Where we had hoped for stories or anecdotes from their experiences on the bus, we were getting mono-syllabic or terse responses, which answered our questions in the most direct way without giving too much information.

The observer effect is a the scientific concept that the behaviour exhibited by a phenomenon will change due to the simple fact of being measured. There’s a similar concept called oberver-expectancy effect where a researcher’s own cognitive biases cause them to influence the participants of an experiment.

This is what I think was happening when we questioned people about their habits, routines, and experiences with the bus. They began to respond with answers they thought were more relevant to us, without oversharing; we began to ask more leading questions about how the wait for the bus was frustrating. We are affecting the responses of the study.

However, in this particular case, we had the luxury of not just asking questions about the system, but to become a part of it. This made a world of difference in the quality of information we collected to work with.

We soon gave up on this process, and decided instead that I would join in with the people actually waiting at the bus stop, while Lisa would go stand by the crosswalk. Immediately, we began to notice a lot more details about the behavior of people. We were also better able to relate with what they were experiencing. The shared experience also made us less “outsiders”, instead we became one of the bus riders, in the same boat as everyone else. Instead of us initiating the conversations, people began to strike up conversation with us to talk about the frustrations of waiting for the bus.

We had a similar experience a second time around (a week later) where Kate and I were sitting at the same bus stop, and got involved in a (fascinating) conversation with an older black woman about her hair, and her impending trip to the salon. Kate was also asked by another woman about the bus-tracking app she was using on the phone. Additionally, we overheard/witnessed several snippets of interactions among others at the bus stop.

From my own history as a regular rider, I feel that the experience of waiting for the bus is one that cannot be reproduced accurately outside of the time and place that it actually takes place in. In talking about it at a later time, the experience is either watered down or exaggerated in its recounting. To truly understand what waiting for the bus is like, one must be present to observe and participate in it as it happens.

While we’re still brainstorming ideas to move forward, the consensus among the team is that the best (and possibly only) way to judge the efficiency of any solution we propose is to prototype it and place it within the actual environment of the bus stop. This will force us to consider realities of context, and give us a means to accurately judge the impact of our design.