An empirical study on the usability of an information navigation aid

Author(s): David Ribeiro Lamas, Jennifer Jerrams-Smith, David Heathcote, Feliz Ribeiro Gouveia
Venue: Proceedings on the 2000 conference on Universal Usability
Date: 2000

Type of Experiement: Controlled Experiment
Sample Size: 40

Quality
3

Pages: 151 - 152
DOI Bookmark: http://doi.acm.org/10.1145/355460.355553

This paper examines the Computer Aided Information Navigation (CAIN)
project's ability to provide adaptive navigation support to increase the World Wide Web's value as a pedagogical tool. CAIN attempts to provide a way of dealing with the massive amount of information currently available on the web. According to the authors, "CAIN provides direct guidance navigation support as a form of nonobtrusive weak hypertext linearization enabling the user to follow a context specific ranked sequence of selected Web pages without ever needing to perform any search or follow any link if they do not wish to."

The authors conduct an experiment to compare CAIN with conventional web tools by comparing them on three measures of performance: comprehension scores, task completion times and user satisfaction. Comprehension is measured by a multiple-choice comprehension questionnaire. Time taken is the measured time to complete the task. And user satisfaction is measured by a like/dislike rating completed at the end of each task. The author states, "Users are either inexperienced in Web use or have little expertise in the target subject topic, or both." The users were required to learn from the World Wide Web in the categories of cryptography and poetry with the using either CAIN or the conventional tools. The authors report their test group as consisting of "of 24 undergraduate students, 12 postgraduate students and 4 staff members. There were 28 males and 12 females in the sample."

The authors found that comprehension scores were 30% greater with CAIN than with the conventional tools. The authors also suggested that task completion time could be reduced without reducing user satisfaction.

0