Page 34 - IJPS-6-2

Internet use in older African Americans

of caregivers, the negative effects of widowhood/widowerhood on daily life may affect widowers more than widows. A
widower living alone at his home may need to spend more time to do housework and take care of himself than he did
before his bereavement, which may reduce his time spent online. By contrast, a widow may need to spend more time
online to gain more social interaction and support. Thus, the effects of correlates may be different between older men and
women. In analyzing internet withdrawal reasons among older adults in Taiwan, Chiu and Liu (2017) found that older
women were more likely to be affected by health factors than older men. Therefore, it is valuable to examine correlates
of internet use among older adults by gender. Furthermore, considering that women live longer than men and, typically,
there are more older women than men in study samples, results may be misleading for older men if they are not separated
in analysis. Thus, in this study, we will investigate correlates among older African American men and women separately.
1.3. Age Differences among Older Adults

Age differences in internet use among older adults have been well confirmed by many previous studies (Anderson and
Perrin, 2017; Chang, McAllister, and McCaslin, 2015). Compared with young-old adults, old-old adults were less likely to
use the Internet. Meanwhile, they had higher risks of experiencing loneliness, cognitive decline, and increased functional
limitations than their young-old counterparts (Cohen-Mansfield, Shmotkin, Blumstein, et al., 2013; Koo, Kõlves, and De
Leo, 2017). Numerous studies have indicated the potential of internet use in reducing loneliness, depression, stimulating
cognitive functioning, and promoting psychological well-being among older adults, especially old-old adults (Sims, Reed,
and Carr, 2017; Fang, Chau, Wong, et al., 2018). Thus, increasing the adoption and use of the Internet among old-
old adults is of great value. However, considering differences between young-old adults and old-old adults, especially
those involving health and mental health conditions, correlates of internet use may be different. Internet use is complex
and requires high cognitive functioning. Thus, different from young-old adults who have better cognitive functioning,
developing the skills necessary to use the Internet may be more challenging for old-old adults. Therefore, it is important
to investigate correlates of internet use by age groups to assist the design and implementation of age-tailored interventions
to increase African American older adults’ internet use.

1.4. Current Study
An overview of the literature has showed several research gaps in the literature involving African American older adults’
internet adoption and utilization. First, most racial or ethnic disparity studies included race or ethnicity as a categorical
variable and compared groups. However, very few studies focused on the African older adult group to explore correlates
of the internet use among this specific population. Second, gender and age differences in internet use among older
adults indicate the necessity of more studies to investigate correlates of different subgroups and conduct more tailored
interventions to increase internet adoption and use. Thus, focusing on African American older adults, the present study
will (1) compare internet use prevalence among older African American by gender and age group and (2) investigate
correlates of internet use among African American older adults by gender and age groups.

2. Method

2.1. Study Sample

Data were drawn from the 2016 Wave of the Health and Retirement Study (HRS), the most recent available data from the
pertinent archive. The HRS is a nationally representative longitudinal panel survey of U.S. individuals aged over 50 and
their spouses. Questions about participants’ health, family structure, retirement, subjective well-being, and lifestyles have
been collected. HRS is funded by the National Institute on Aging (grant number NIA U01AG009740) and is conducted
by the University of Michigan. The Institutional Review Board at the University of Michigan and the National Institution
on Aging provided ethical approval for the HRS.
Two separate data files – 2016 HRS Core (Final) and RAND HRS Longitudinal File 2016 – were merged based on
key values in the current study. internet use and perceived daily discrimination were obtained from the Left-Behind
Questionnaire from the HRS core data. All other demographic, socioeconomic, health, and mental health variables were
drawn from the RAND HRS data files. In the 2016 Wave, 20,912 adults participated in the survey. First, since the
Left-Behind Questionnaire was assigned to a rotating (random) 50% of participants who completed the enhanced face-
to-face interview, 10,674 participants who were ineligible for the Left-Behind Questionnaire were removed from this
study. Second, among the total eligible sample, 3882 participants who did not return or complete the questionnaire and
32 questionnaires completed by those who were not the assigned respondents were excluded from this study. Third, after

28 International Journal of Population Studies | 2020, Volume 6, Issue 2