Page 36 - IJPS-6-2

Internet use in older African Americans

2.2.5. Demographic factors
Previous studies have found that age, gender, and living arrangements were important correlates of internet use among
older adults (Ang, Lim, and Malhotra, 2020; Chang, McAllister, and McCaslin, 2015; Choi and DiNitto, 2013; Choi, Kim,
Chipalo, et al., 2020; Elliot, Mooney, Douthit, et al., 2014). The Pew Research Center Survey indicated that internet use
among older adults dropped dramatically around 75 years of age (Anderson and Perrin, 2017). Consequently, in this study,
age was categorized into a binary variable (0 <75 years, young-old; 1 = 75 years or older, old-old). Gender (0 = men,
1 = women) and living arrangements (0 = living alone, 1 = living with others) were dichotomous variables.
2.3. Analytic Strategies

First, descriptive statistics were used to better understand the characteristics of the African American older adults in
this study and their internet use status. Next, bivariate analysis of internet use was conducted to investigate gender and
age differences in internet use. Third, considering that the dependent variable, internet use, was measured as an ordinal
variable, ordinal logistic regressions were performed in this study. Two sets of sequential ordinal logistic regressions
were conducted separately by gender and age. Four models were performed for both sets: in Model 1, only demographic
factors were added; in Model 2, both demographic and socioeconomic factors were included; in Model 3, demographic,
socioeconomic, and health-related factors were assessed; in Model 4, demographic, socioeconomic, health-related, and
mental health-related factors were analyzed. All analyses were performed using Stata 14.2.

3. Results

3.1. Sample Characteristics

Table 1 shows descriptive characteristics of the African American older adults in the study sample. About 18% of the
sample was aged 75 years or older, 63% were women and about 29% were living alone. The participants had about
13 years of education on average (SD = 2.73), or just above high school completion. Less than half of the participants
were married/partnered (45%) or retired (46%), and about 22% were living in poverty. As for their health, the mean
self-rated health was 3 (SD = 0.99), and on average participants had 2 chronic illness (SD = 1.39). Participants had
very low difficulties in ADL (M = 0.41, SD = 0.98) and IADL (M = 0.27, SD = 0.77). Regarding mental health-related
factors, African American older adults had relatively low levels of depression (M = 1.68, SD = 2.00) and perceived

Table 1. Sample characteristics of African American Older adults (n=1117).
Mean SD n %
Demographic
Old-old 200 17.9
Women 707 63.3
Living alone 327 29.3
Socioeconomic
Years of education (0-17) 12.84 2.728
Married or partnered 506 45.3
Retired 509 45.6
Living in poverty 242 21.7
Health-related
Self-rated health (1-5) 2.94 0.994
Number of chronic illness (0-7) 2.19 1.390
Difficulties in activities of daily living (0-5) 0.41 0.981
Difficulties in instrumental activities of daily living (0-5) 0.27 0.767
Mental health-related
Depression (0-8) 1.68 2.003
Discrimination (1-6) 1.75 0.838
Cognitive functioning (0-35) 20.28 4.613

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