A model of computer-mediated social support among older adults

Abstract

Currently, older adults are among the fastest growing online users, and the Internet/e-mail could serve as a new support system for these individuals. The purpose of this study was to test "A Model of Computer-Mediated Social Support Among Older Adults" that explains relationships among a computer-mediated social network (CMSN), functional social support from that network (CMFSS), and psychological well-being of older adults. This was a one-time single group descriptive study employing a Web survey. The questionnaire was posted on the SeniorNet Website from April 5 to June 9, 2002. Participants were eligible if they were: (1) age 55 or older; (2) able to use the Internet/e-mail independently; and (3) currently residing in a community setting. The data were directly populated to an ACCESSRTM database. The data then was imported into SPSS for data screening, and then into AMOS 4.0 for the SEM analysis. A two-step approach (measurement model test first, then full model test) was employed in this model testing. Validity of the overall model was examined employing the cross-validation method. A total of 809 SeniorNet users participated in this study. The majority of participants were white (751, 92.8%) with a mean age of 67.8. About two-thirds of participants (511, 64%) were female. The majority (641, 80.3%) reported at least some college education. Model testing showed that, overall, the proposed model fit the sample data. The fit indices were: chi2/df, 2.86 (chi 2 = 1358.39, df = 475, p < .001), GFI, .90; RMSEA, .05; CFI, .88; NFI, .82. Specific findings demonstrated that computer time, computer knowledge, age, gender, and physical health directly influenced CMSN, and CMSN had a positive impact on CMFSS. Computer-mediated social support, however, did not have a significant impact on psychological well-being. The supported relationships between CMSN and CMFSS suggest that health care professionals may use Internet/e-mail to provide important health-related support (e.g., information and support groups) to older adults. The non-significant relationships could have been due to other confounding variables or measurement issues in psychological well-being; this calls for further research.