Recent Content
Why you only need to test with five users (explained) : March 8, 2010
For finding usability problems with an interface, testing with five users is fine to find problems that affect 31% to 100% of all users. If a problem is more elusive (affects fewer than 31% of users) then you need to increase your sample size. This sample size does not apply to comparing designs or generating a precise estimate of completion rates or task-times.
Can we trust data from professional usability test-takers? : February 9, 2010
In a comparative test, satisfaction scores and completion rates from professional usability test-takers were nearly identical to lab-based users. However, time on task data differed significantly and showed much higher variability. For testing websites intended for a general audience the use of professional testers appears to provide mostly reliable data quickly and for a fraction of the price.
Does better usability increase customer loyalty? : January 7, 2010
I examined the relationship between customer loyalty as measured by the Net Promoter Score (NPS) and System Usability Scale (SUS) questionnaire from several usability tests. I found that perceptions of usability account for about 1/3 of the changes in customer loyalty. Increasing your usability will lead to increased loyalty.
How to conduct a Quantitative Usability Test : December 8, 2009
Do you know how to measure usability? Do you have questions about the benefits and process for conducting a quantitative usability test? I've assembled answers to the 72 most common questions that arise from measuring usability. In this 64 page report I provide concrete examples and plenty of data from a dataset of 120 usability tests, the latest usability research and my decade of experience conducting quantitative usability tests.
Is there a difference in usability data from remote unmoderated tests and lab-based tests? : December 8, 2009
Is it possible to get the same data from lab-based tests by having users test themselves? Unmoderated testing appears to provide a cost effective alternative for gathering a lot more usability data with considerably less effort. Additional time is required to filter invalid data such as unrealistically short task times.
How much is a PhD Worth? : November 5, 2009
Does a full-time PhD pay off in the Usability Profession? The UPA Salary data from 2009 and 2005 is analyzed. It shows that while a PhD may open doors, being out of the work-force for five years is an opportunity cost that is unlikely to made up for.
Do users fail a task and still rate it as easy? : October 9, 2009
Only 14% of users who fail a task rate it at maximum level of satisfaction. In general there is an 80/20 rule of satisfaction and completion rates: 80% of users who rate at the maximum level of satisfaction will pass and 80% of users who rate at the minimum satisfaction level will fail the task.
Usability Statistics Package Expanded : October 1, 2009
Calculate the common statistical tests, sample sizes and use some advanced statistical tests on usability data. This calculator package contains everything the Usability Statistics Package contains, but is expanded to included sample sizes calculators for margins of error and power calculations for comparing two applications. It also contains ANOVA and Chi-Square tests to compare multiple means or completion rates. This package takes the guess work out of what tests to perform on your usability data.
Margins of Error in Usability Tests : August 6, 2009
How many users will complete the task and how long will it take them? If you need to benchmark an interface, then a summative usability test is one way to answer these questions. Summative tests are the gold-standard for usability measurement. But just how precise are the metrics?
Compare 2 Small Sample Completion Rates (Fisher Exact Test) : June 5, 2009
Used for comparing 2 small sample binary completion rates (it uses a statistical test called the Fisher Exact Test)
Task Times in Formative Usability Tests : June 6, 2008
Time-on-task can be used as a valuable diagnosis and comparative tool during formative evaluations.
