Applicable stages: test, and deployment.
Personnel needed for the evaluation: Usability experts: 1 Software developers: 0 Users: 6	Usability issues covered: Effectiveness: Yes Efficiency: No Satisfaction: Yes
Can be conducted remotely: No	Can obtain quantitative data: No

 Overview

This is a data collecting technique where about 6 to 9 users are brought together to discuss issues relating to the system. A human factors engineer plays the role of a moderator, who needs to prepare the list of issues to be discussed beforehand and seek to gather the needed information from the discussion. This can capture spontaneous user reactions and ideas that evolve in the dynamic group process.

The general procedures for conducting a focus groups study is:

• Locate representative users (typically 6 to 9 per focus group) who are willing to participate.
• Select a moderator.
• Prepare a list of issues to be discussed and goals for the type of information to gather.
• Keep the discussion on track without inhibiting the free flow of ideas and comment.
• Ensure that all participants get to contribute to the discussion. Guard against having a single participant's opinion dominate the discussion.
• Have the discussion feel free-flowing and relatively unstructured to the participants, but try to follow a prepanned script.
• Write a summary of the prevailing mood and critical comments of the session, including representative quotes.

The following is a list of things to be considered when conducting a focus groups study:

• Consider having more than one focus group, since the outcome of a single session may not be representative and a single discussion may have focused on a subset of the issues or minor aspects of the system.
• The moderator needs to be skilled in group facilitation and communication to make a focus group successful. It is not as simple as preparing questions, because the moderator needs to facilitate and guide discussion in real time.
• The data colleced tend to have low validity and are very difficult to analyze because of their unstructured and free-flowing nature.
• Computer conferencing or electronic mail networks or bulletin boards may be an alternative way of simulating the focus group approach. However, their limitation is that the people who are responding are probably not representative users, but rather expert users.

1. J. Nielsen "Usability Engineering" pp.214-216, Academic Press, 1993.