Recent studies indicate that altitudes toward computers are improved in old
er adults following brief training sessions (G. Jay & S. L. Willis, 1992; R
. W. Morrell, D. C. Park, C. B. Mayhorn, & R. V. Echt, 1996a; E. Zandri & N
. Charness, 1989). In the present research, the utility of the attitude con
struct as a predictor of computer use is explored. Thirty-nine older adults
were trained to use the ELDERCOMM electronic bulletin board system (R. W.
Morrell, D. C. Park, C. B. Mayhorn, & R. V. Echt, 1996b), and they were the
n given the opportunity to use the system as often as they liked over a 2-m
onth participation interval. Participants in the study had more positive at
titudes toward computers than an unselected group of older adults, and the
most positive attitudes were found in the group who returned to use the bul
letin board frequently during the participation interval. However, the most
important predictor of continued use of the system is success at initial t
raining. Age, education, and income did not predict continued use of the sy
stem, despite strong effects of these variables on computer adoption in nat
ional samples (U.S. Census Bureau, 1984, 1989, 1993). It is suggested that
these variables reflect opportunity rather than desire to use computers. As
in previous research, brief training serves to improve attitudes toward co
mputers. No effects of computer training on psychosocial and emotional adju
stment variables such as depression, anxiety, and self-esteem are found. Th
ese results are discussed with respect to efforts designed to induce older
adults to use computers with greater frequency.