What the scientific method taught me about user research

Jun 27 2014

A screen from the prototype we showed users this week.

A screen from the prototype we showed users this week.

I am a chemistry and photojournalism major, and I find that it is very rare for my science courses to play a large role in our projects in Reese News Lab.

But this week my team dove into the scientific method in order to test the desirability of our Web interface. Our team was tasked with creating a Web interface to connect the community with local nonprofit organizations. In just five days, we developed a prototype, allowed users to interact with it, and made alterations all while implementing validated learning, a version of the traditional scientific method. Here’s how we did it:

Step 1: Background research
Over the past few weeks, our team has talked to many nonprofit organizations to determine what they actually need and what they would use. At the same time, we talked to volunteers and community members about how they currently interact with nonprofits to determine what we can do to increase these interactions.

Step 2: Start with a hypothesis
Our background research indicated that users may be more likely to donate to a nonprofit if a specific need of the organization – such as canned goods or toys – is presented. We decided to go out into the community and present random people with a list of five needs of charities around Chapel Hill.

Step 3: Test the hypothesis
We set out with the list of needs and clipboards. We began asking people if they would help out local charities by donating one of these items. The needs were not linked to a specific charity. We tried to make the interaction as real as possible because we didn’t want a hypothetical situation to lead to hypothetical answers. The majority of the people we asked said that they would be willing to donate and gave us their email address without asking which charity was asking for the donations. Only one person inquired about the organization.

Step 4: Lather, rinse, repeat
We then realized that we have to test the other side of the hypothesis in order to determine if we happened to pick the most charitable people on campus. We embarked again into the community and presented them with a list of charities in the area. When we asked if they would be willing to donate, only a few people agreed to give us their contact information.

Step 5: Create the prototype
With this knowledge in hand, we went back to the drawing board to create an interface that put the needs first. We created a prototype on paper and then on Microsoft PowerPoint that would replicate the experience of a website.

Moving forward
Many scientists may argue that our methods were flawed because of lack of controls or differences in the test subjects’ environments. However, we were able to apply the process of developing a specific hypothesis to test a specific aspect of our interface. The key lesson we learned this week is to always define the reason behind the test and what we are hoping to learn from the test prior to asking the community for feedback.


Leave a Reply

Your email address will not be published. Required fields are marked *