By the year 2020, it is projected that 30% of the United States population will be comprised of people age 65 and older (Administration on Aging, 2004). Individuals over age 65 will continue to constitute a larger proportion of the total population because people are living longer and healthier lives. With older adults living longer, this senior population leads very active lives and often has great interest in modern technology such as the Internet (Nielsen, 2002). Given the use of computers in the workplace and homes and the increase in the number of older adults in the next 20 years, the use of computers by older adults is a significant issue that should be addressed (Czaja, 1996).

Computer tasks involve cognitive skills that may be challenging for older adults. Most of the literature suggests that cognitive skills decline as individuals age (Baddeley, 1981; Foos, 1989; Salthouse, 1996; Welford, 1985).

Decrements in working memory could place older adults at a disadvantage when performing computer-interactive tasks. To increase the success of older workers' performance with computer technology, web-based training programs designed to accommodate age-related cognitive declines may be an effective avenue to deliver computer training to younger and older adults. The objectives of this study were to explore computer self-efficacy differences between younger adults (18-24 years old) and older adults (65 and older) and to determine if the design of a web-based training system affects individual performance when completing a computer task. Four hypotheses were tested in this study: older adult-centered web-based training program would yield better performance for both younger and older adults; older adults would have lower computer self-efficacy than younger adults; participants with higher computer self-efficacy would perform better on a computer task; and participants with higher usability ratings would perform better on the computer task. Results of the study did not directly support any of the hypotheses. However, age-related differences were evident in training time, task completion time, performance score, and usability ratings. The older adult participants had longer training and task completion times, lower performance scores, and higher usability ratings. Results were discussed in terms of limitations and implications of older-adult centered training programs.