Developed in the late 1990s, the modernization theory of aging posited that older adults were in danger of losing control and power over their lives because they could not keep up with technological progress. Early concerns about the age-related digital divide focused more on the access to technology; however, the age-related digital divide is a complex and multidimensional phenomenon. Previous works on technology use are not without substantial inconsistencies, and the research findings on antecedents and consequences of technology use remain especially equivocal. Without refining technology construct, inconsistent findings could hinder the understanding of different associations among distinct forms of technology used by older adults.

This dissertation consists of three studies which examined predictors of refined technology construct health-related information technology (HIT), work-related information technology (WIT), communications technology (CT), entertainment technology (ET) and patterns of technology use in older populations, using the data from the most recent wave (2011) of the Wisconsin Longitudinal Study (WLS). I further explored patterns of technology use in older populations that have been overlooked in previous studies.

Results from the first study showed that sets of factors (i.e., individual characteristics, social roles, and personality traits) differently predicted each type of technology use. Older women were more likely to use HIT and CT, but less likely to use WIT. However, there were no gender differences in ET use. Older adults with higher subjective health predicted the use of WIT, CT, and ET, but not HIT. Among the Big Five personality traits, openness predicted all types of technology use, higher agreeableness was associated with less use of both HIT and WIT, and less conscientiousness was associated with less use of ET.

In the second study, I applied latent class analysis to find the best fitting model to explain the patterns of technology use in older populations. It yielded a 3-class model, where each class was identified as multi-users, selective users, and non-users. Sets of factors (i.e., individual characteristics, social roles, and personality traits) predicted each class membership differently. Multi-users were more likely to be women, younger, married, in families with higher household income, in better subjective health, more education, and higher openness than non-users. Selective users were more likely to be employed, married, in better subjective health, more education, and have higher agreeableness and openness than non-users.

In the third study, I examined associations between patterns of technology usage and multidimensional psychological well-being of older adults. Results from ordinary least squares regression (OLS) models showed that selective users, but not non-users, had lower levels of depressive symptoms compared to multi-users. Non-users reported lower levels of psychological well-being compared to multi-users in all six sub-domains (autonomy, environmental mastery, personal growth, positive relations with others, purpose in life, & self-acceptance) of Ryff’s psychological well-being scales. Also, selective users showed a lower level of personal growth compared to multi-users.

Taken together, the findings from this dissertation contribute to the literature examining technology use of older adults and its antecedents and outcomes. In particular, refined technology constructs demonstrate diverse aspects of technological use in older populations, and the explored typology provides a framework to translate findings into intervention programs which will consider multidimensionality of older populations in technology use.