While I was in graduate school at San Diego State University in 1999 I studied usability for my thesis. This short tutorial is from a literature review I did on the Think Aloud Method of usability testing any screen based media.
How to test quickly test any screen media using the Think Aloud Method
“Our mind is not like a brilliantly lit and perfectly ordered room; it is much more like an encumbered garret inhabited by moths born and grown up in half lights: our thoughts; the moment we open the door to see them better the drab little butterflies vanish.” (Dimnet, 1928)
“The Think Aloud Method consists of asking people to think aloud while solving a problem and analyzing the resulting verbal protocols.” (Someren, 1994)
The Think Aloud Method is simply a way of getting people to tell you what they think while they use something. In our case that something is a Website. The Think Aloud Method is specifically good for flushing out a user’s mental model which gives the researcher a window into the user’s thought process when they use the site. Because most Web surfing/mobile use happens while the user is alone, or at least socially isolates themselves briefly, this method is particularly good for Website and mobile application testing.
Our job is to try and catch those butterflies
- Subject is coached that this is a test NOT of them, but of a site they are helping to simulate what people might think while they surf the site alone.
- Subject thinks aloud (talking outloud) while performing a simple and clearly defined task.
- Subject does not analyze thoughts, just verbalizes them. Literally, “speaking their mind”
- Talk is continuous.
- Researcher bites tongue and takes notes or video records session and/or screen.
- Research only prompts with neutral phrases like “keep talking.”
- Do not use questions such as: “what do you think of that?” which will force the user to make up an answer.
The Key Attributes
When to use
- To find major usability problems with a few subjects
- Finding the “why” of problems
- With subject matter experts (SMEs) to find out what they don’t know they know.
- Alpha and beta testing for mid-project design changes
When not to use
- Tasks that involve heavy cognitive loads that would interfere with talking out loud
- For long tests
- For opinions
- With very young kids
- Reveals mental models
- Reveals cognitive processes
- Flexible for location, equipment, recording
- Unnatural for some users and kids
- Data can be hard to analyze
- Users can rationalize and distort data
- Someren, W. M., Barnard, F. Y., & Sandberg, A. C. J. (1994). The think aloud method. A practical guide to modeling cognitive process. London: Harcourt Brace & Company.
- Dimnet, E. (1928). The art of thinking. New York, Simon and Shuster Inc.