Trends in Internet Use Among Older Adults in the United States,2011

Abstract

The present study examined the trends in internet use among older adults (60+ years) in the United States from 2011 to 2016 by considering not only the entire study sample as a whole but also specific subgroups by age, gender, race/ethnicity, and education. Using data from 107,500 older adults in the California Health Interview Survey between 2011 and 2016, the significance of linear trends was tested by including the survey year as a continuous variable in the logistic regression. Findings confirm the general knowledge that an increased number of the older U.S. population used the internet over the period of 2011–2016. Furthermore, closer inspection indicates that this upward trend does not apply to all, but specifically to those with advanced age, male gender, racial/ethnic minority background, and low educational attainment. Findings provide implications for identifying and prioritizing a target group for internet technology training.

Keywords

older adults trends in internet use digital divide

Introduction

The use of the internet among older adults has become integral, empowering them to enjoy a productive and independent life. A sizable body of literature demonstrates numerous benefits that internet use brings to older adults, including enhanced convenience in life, connectedness to information and social networks, and physical, mental, and cognitive health and well-being (Khalaila & Vitman-Schorr, 2018 ; Pew Research Center, 2018; Wagner et al., 2010).

However, the digital gap or disparity in access to the internet still persists among certain segments of the older population in the United States. In particular, age, gender, race/ethnicity, and socioeconomic status (SES) have been consistently reported to be major sources of such disparities. The rate of internet use is notably higher among the young-old compared with their counterparts with advanced age (Cresci et al., 2010; Pew Research Center, 2017; Seifert et al., 2017; Van Deursen & Van Dijk, 2014). A lower rate of internet use in older female adults has consistently been documented (König et al., 2018). As in the case of other technology adoptions, older African Americans and Hispanics were less likely to be the users of the internet than older White people (Hunsaker & Hargittai, 2018; Yoon et al., 2020). In addition, low SES, often indicated by fewer years in education, has been shown to be a critical barrier to older adults’ use of the internet (Lin et al., 2015; Werner et al., 2011).

Despite well-documented disparities in access to the internet among older adults, there is a dearth of information on the trends in internet use within these subgroups. To address this knowledge gap, the goal of the present study was to examine the trends in internet use, not only in the older population in general but also in a specific subgroup by age, gender, race/ethnicity, and education. Our selection of the sociodemographic sources of disparities was guided by previous research reporting that older adults with advanced age, female gender, racial/ethnic minority background, and low educational attainment are less likely to use the internet and related technologies (König et al., 2018; Pew Research Center, 2017). We hypothesized that (a) the rate of internet use by older U.S. adults would increase between 2011 and 2016 and (b) there would be disparities in such trends among subgroups with advanced age, female gender, racial and ethnic minority background, and low educational attainment. Understanding of the within-group trends will facilitate the identification of subgroups that lag behind and help prioritize efforts to address the digital divide in the older population.

Method

Sample

The California Health Interview Survey (CHIS) is a population-based random-digit dial telephone survey of California’s non-institutionalized population and employed both disproportional stratified sampling and multiple frame sampling methods to increase the representativeness of racial and ethnic groups in California. To capture the rich diversity of the Californian population, languages used in the interviews included English, Spanish, Chinese (Mandarin and Cantonese), Vietnamese, and Korean. Detailed information about the CHIS is available elsewhere (CHIS, 2017).

The present study used the pooled cross-sectional design of 2011, 2012, 2013, 2014, 2015, and 2016 CHIS adult data sets. Since 2011, a total of 227,948 respondents have completed the CHIS survey; however, the current study restricted analyses to older adults aged 60 years and over who self-identified their race or ethnicity as non-Hispanic White, African American, Latino, or Asian American between 2011 and 2016. Other racial/ethnic groups (e.g., American Indian/Alaskan Native, Pacific Islander, and other single/multiple race) were excluded due to their small sample sizes. Therefore, the sample consisted of older adult respondents (N = 107,500), with subgroups of non-Hispanic White (n = 79,424), African American (n = 4,973), Asian American (n = 10,229), and Latino (n = 12,874).

Measures

Outcome variable

Internet use was assessed with a single-item question: “Have you ever used the internet?” Answers were coded as 0 (no) and 1 (yes).

Subgroups of interest

Participants were grouped into young-old (60–74 years), old-old (75–84 years), and oldest-old (85+ years). Gender was coded as a binary variable. Groups for race/ethnicity include non-Hispanic Whites, African Americans, Asian Americans, and Latinos. Educational level was grouped into the following four categories: less than high school graduation, high school graduation, some college, and college graduation or above.

Control variables

Variables controlled for the trend analyses were marital status (0 = married or living with partner; 1 = widowed, separated, or divorced; 2 = never married) and self-rated health (0 = excellent/very good/good; 1 = fair/poor).

Analytic Strategy

Descriptive characteristics of the full sample were examined. First, we tested the significance of the linear trends in internet use over the period of 2011–2016 in the full sample. Then, we examined the trends in internet use within each subgroup of interest by age, gender, race/ethnicity, and education, controlling for the effects of the covariates (e.g., marital status and self-rated health). Following the guidelines of recent trend studies (e.g., Centers for Disease Control and Prevention, 2016), the significance of linear trends was tested by including the survey year as a continuous variable in the logistic regression. The pooled dataset was weighted to account for the stratified sampling design as recommended (CHIS, 2017), and all analyses were conducted in STATA 14. The use of publicly available data from the CHIS has been approved by the institutional review board of the Texas State University.

Results

Characteristics of the Sample

Sample characteristics of the full sample are summarized in Table 1. Approximately two thirds of the full sample (67.3%) were internet users. More than 64% were in the young-old group aged between 60 and 74 years. The sample includes 79,424 non-Hispanic Whites (73.9%), 4,973 African Americans (4.6%), 10,229 Asian Americans (9.5%), and 12,874 Latinos (12.0%). About 60% of the sample were female, 9.9% had an educational level of less than high school graduation, and about a half (49.8%) were married or living with a partner. With regard to health, 25% reported that their health was either “fair” or “poor.”

Table 1.

Descriptive Characteristics of the Full Sample of Older Adults: CHIS 2011–2016.

Variables	Total (%)
Outcome variable
Internet use
Yes	67.3
No	32.7
Subgroups of interest
Age
Young-old (60–74)	64.1
Old-old (75–84)	25.6
Oldest-old (85+)	10.3
Gender
Male	40.1
Female	59.9
Race/ethnicity
Whites	73.9
African Americans	4.6
Asians	9.5
Hispanics	12.0
Education
Less than high school graduation	9.9
High school graduation	20.3
Some college	28.7
College graduation or above	41.1
Control variables
Marital status
Married/living with partner	49.8
Widowed/separated/divorced	43.3
Never married	6.9
General health condition
Excellent/very good/good	75.0
Fair/poor	25.0

Note. CHIS = California Health Interview Survey.

Trends in Internet Use Among Older Adults

Table 2 shows the linear trends in internet use over the period of 2011–2016 in the full sample and each subgroup of interest. As shown in Figure 1 and Table 2, the prevalence estimates for internet use among the full sample increased significantly from 60.2% in 2011 to 66.4% in 2016 (adjusted odds ratio [AOR] = 1.08, 95% confidence interval [CI] = [1.04, 1.13], p < .001). When the trends were tested separately for each subgroup, a statistical significance in the increase of the trend was observed in the groups with young-old age, old-old age, female gender, and high school education or above. With respect to race/ethnicity, White older adults’ internet use increased significantly from 66.3% in 2011 to 72.5% in 2016 (AOR = 1.10, 95% CI = [1.06, 1.15], p < .001), reflecting a 9.4% proportional increase in the prevalence of internet use over 6 years. However, during the same time period, no statistically significant changes in trends were observed among other racial/ethnic minority groups.

Table 2.

Test of Significance for Trends in Internet Use among Full Sample and Subgroups by Age, Gender, Race/Ethnicity, and Education: CHIS 2011–2016.

Variables	AOR	95% CI
Full sample	1.08***	[1.04, 1.13]
Subgroups of interest Age
Young-old (60–74 years)	1.09**	[1.04, 1.14]
Old-old (75–84 years)	1.08**	[1.02, 1.15]
Oldest-old (85+ years) Gender	1.09	[0.99, 1.20]
Male	1.03	[0.98, 1.09]
Female Race/ethnicity	1.12***	[1.06, 1.18]
Whites	1.10***	[1.06, 1.15]
African Americans	1.03	[0.96, 1.11]
Asians	1.13	[0.99, 1.27]
Latinos Education	1.07	[0.96, 1.19]
Less than high school graduation	1.10	[0.97, 1.25]
High school graduation	1.08**	[1.02, 1.14]
Some college	1.10**	[1.03, 1.18]
College graduation or above	1.08**	[1.04, 1.13]

Note. The AORs were calculated for the full study sample and each subgroup separately. For each subgroup, other sociodemographic and health variables were controlled. Significant odds ratios with a value of greater than 1.00 reflect an increase in trend, and significant odds ratios with a value of less than 1.00 reflect a decrease in trend. CHIS = California Health Interview Survey; AOR = adjusted odds ratio; CI = confidence interval. **p < .01. ***p < .001.

Figure 1.

Prevalence estimates of internet use among older adults in the United States: CHIS 2011–2016.

Discussion

In response to the integral role of internet use in empowering older adults to enjoy a productive and independent life and the rapid increase of internet use in the older population (Pew Research Center, 2017), the present study examined the trends in internet use among older adults in the United States in the period of 2011–2016, considering population in both general and specific subgroups. As demonstrated in other national studies (e.g., Hong & Cho, 2016; Pew Research Center, 2017), the prevalence estimates for internet use among older adults in the full data increased significantly from 60.2% in 2011 to 66.4% in 2016. When it comes to specific sociodemographic subgroups, the significant increases in internet use between 2011 and 2016 were observed among those who were young-old, old-old, female, and non-Hispanic Whites, and those with high school graduation or above. The findings are in accordance with previous studies showing the important role of older adults’ sociodemographic characteristics in predicting internet use (Werner et al., 2011; Yoon et al., 2020). Consistent with the literature, the findings of the current study showed that a younger age was a strong predictor of internet use, indicating that internet use among oldest-old adults aged 85 years and older in the United States has not substantially changed between 2011 and 2016. While the current study found that men were more likely than women to be internet users (70.6% vs. 65.3%), significant changes in trends were observed only among women. This may imply that the digital divide between men and women has narrowed over the period of 2011–2016. Educational attainment, as a proxy for SES status, has also been found to be a strong indicator of internet use among older adults. While the vast majority of older adults with more than high school graduation have been increasingly using the internet, no significant changes in trends were observed among older adults without a high school diploma. Another finding confirmed in the study is that race/ethnicity is a source of a digital divide. While we have observed an increase in White older adults over the period of 2011–2016 using the internet, such an increase did not necessarily apply to older racial/ethnic minorities. Therefore, our findings confirm that the multiple benefits of internet use are not shared by all members of the older U.S. population. In particular, previous studies found that low SES and racial/ethnic minority status are not only independent predictors of internet use but their intersectionality also plays a critical role (Yoon et al., 2020). In other words, while the effect of SES on internet usage is robust for White older adults, it is an even stronger predictor for racial/ethnic minority older adults. Such information should be incorporated in intervention efforts for internet training/education targeted toward racial/ethnic minority older adults with low SES.

Since 1999, when the U.S. Department of Commerce created several initiatives to provide technology resources to under-served communities and groups, several programs were designed to narrow the digital divide, such as installing computers and the internet services in public schools, libraries, and health care facilities in impoverished areas (National Telecommunications and Information Administration, 1999). However, the current findings show that age, race/ethnicity, and low education still operate as a barrier to older adults’ access to the internet. That is, older adults with advanced age, racial/ethnic minority background, and less education still lag behind the trend and the associated benefits of technology. In particular, given the importance of the internet as a means of sharing information, and given older adults’ greater need for health information, older adults are in a position to capitalize on the use of technologies by accessing web-based health-related information. Studies report that older individuals who use the internet to seek health-related information experienced improved outcomes with respect to their knowledge of health issues, health communication with medical professionals, and appropriate use of health services (Hong & Cho, 2016). In addition, numerous studies report other benefits for older adults of using the internet such as improving interpersonal interactions, promoting better cognitive functioning, and enhancing their experience of control and independence (Anderson, 2017; Jackson et al., 2010; Shapira et al., 2007). Nonetheless, the findings of the current study highlight that significant variations in internet use among different sociodemographic subgroups of older adults persisted. Along with the general consensus that older adults with advanced age, racial/ethnic background, and low SES have limited opportunity to access the internet and are less likely to adopt new technology (Pew Research Center, 2017), findings call attention to those identified subgroups of older adults in an effort to design and implement technology education programs.

Apart from the findings of the current study, among the major problems causing older adults to avoid using the internet or minimize their use of it are psychological aspects such as technology-use-related anxiety and perceived competence. Findings of recent intervention studies designed to improve technology literacy also highlight the importance of the improvement in computer confidence and self-efficacy and of reduction in computer anxiety among older adults (Doh et al., 2015). It has been speculated that older adults with positive psychological constructs may be more willing to embrace new challenges (e.g., the use of internet and related technologies). Therefore, the findings of the current study provide implications for the importance of age-specific and culturally appropriate internet technology training that meets the psychological and technical needs of diverse groups of older adults to increase their use of the internet. In addition, as older adults with low income are typically unable to afford to have computer equipment and internet access at home, other factors which can play a significant role in facilitating older adults’ technology learning should be considered to ensure successful internet use among older adults, such as providing public facilities where training and internet use take place and offering technology equipment to individuals in need.

Some limitations to the present study should be noted. First, the present study used the pooled CHIS data of multiple years to analyze the trends, but all data are cross-sectional and geographically restricted. Therefore, the present study is limited for drawing causal inferences and generalizing the findings to the larger population of older adults in other geographic settings. Future studies should revisit the topic with a representative sample on a national level and include a longitudinal approach to examine the age-related trends in the pattern of internet use to facilitate generalization and causal inference. Second, it should be noted that internet use was assessed with a single-item question: “Have you ever used the internet?” with a single yes/no item. The question may imply internet use at any time during the survey participants’ lifespan, which does not specify the current internet use during the current age range (60 years and older). This limitation is inherent to the secondary analysis of data. In addition, the way in which the internet was accessed should be considered. Participants may have accessed the internet directly, or it may be that a friend or family member navigated the internet for them. As the dimensions of internet use are diverse, future research should incorporate a more refined measure reflecting the subject’s current internet use and embracing diverse purposes of online activities and methods of accessing the internet. Furthermore, the quality of internet use experience (e.g., where the internet was accessed, what types of information were retrieved, satisfaction, and quality of information) should be examined in future studies. Third, the current study focused on trends in internet use among both all older adults and subgroups, but identifying the actual barriers (e.g., functional impairment—particularly visual and auditory, lack of exposure, or psychological barriers) to internet use can be the logical next step in eliminating the digital divide for the identified target population. Finally, although the model estimated in the current investigation accounted for the important roles of sociodemographic factors of internet use, there might still be other important factors (e.g., well-being, region, income, presence of a computer at home, the possession of a smartphone) affecting internet use, which should be considered in future studies.

Despite these limitations, responding to an important role of access to internet technology in improving well-being, the current study identifies specific groups to be prioritized in efforts to close this digital divide between older adults. Findings also provide implications for culturally appropriate internet technology training for the diverse groups of the target population.

Footnotes

Authors’ Note

This study has been approved by the institutional review board of the Texas State University (with the Approval No. 6627).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Hyunwoo Yoon

Seoyoun Kim

References

Anderson

(2017). Getting connected: Older Americans embrace technology to enhance their lives. AARP Research.

California Health Interview Survey. (2017). CHIS 2016 adult survey. UCLA Center for Health Policy Research.

Centers for Disease Control and Prevention. (2016). Conducting trend analyses of YRBS data. https://www.cdc.gov/healthyyouth/data/yrbs/pdf/2015/2015_yrbs_conducting_trend_analyses.pdf

Cresci

M. K.

Yarandi

H. N.

Morrell

R. W.

(2010). Pro-nets versus no-nets: Differences in urban older adults’ predilections for internet use. Educational Gerontology, 36(6), 500–520.

Doh

Schmidt

L. I.

Herbolsheimer

Jokisch

M. R.

Wahl

H. W.

(2015). Patterns of modern ICT use among “senior technology experts”: The role of demographic variables, subjective beliefs and attitudes. In Zhou

Salvendy

(Eds.), Human aspects of IT for the aged population. Design for aging (pp. 177–188). Springer.

Hong

Y. A.

Cho

(2016). Has the digital health divide widened? Trends of health-related internet use among older adults from 2003 to 2011. Journals of Gerontology Series B: Psychological Sciences & Social Sciences, 72(5), 856–863.

Hunsaker

Hargittai

(2018). A review of Internet use among older adults. New Media & Society, 20(10), 3937–3954.

Jackson

L. A.

von Eye

Fitzgerald

H. E.

Zhao

Witt

E. A.

(2010). Self-concept, self-esteem, gender, race and information technology use. Computers in Human Behavior, 26(3), 323–328.

Khalaila

Vitman-Schorr

(2018). Internet use, social networks, loneliness, and quality of life among adults aged 50 and older: Mediating and moderating effects. Quality of Life Research, 27(2), 479–489.

10.

König

Seifert

Doh

(2018). Internet use among older Europeans: An analysis based on SHARE data. Universal Access in the Information Society, 17(3), 621–633.

11.

Lin

C. A.

Atkin

D. J.

Cappotto

Davis

Dean

Eisenbaum

Vidican

(2015). Ethnicity, digital divides and uses of the Internet for health information. Computers in Human Behavior, 51, 216–223.

12.

National Telecommunications and Information Administration. (1999). Government programs designed to close the divide, fact sheet. https://www.ntia.doc.gov/legacy/ntiahome/fttn99/contents.html

13.

Pew Research Center. (2017). Tech adoption climbs among older adults. https://www.pewinternet.org/2017/05/17/tech-adoption-climbs-among-older-adults/

14.

Pew Research Center. (2018). Social media use continues to rise in developing countries but plateaus across developed ones. https://www.pewresearch.org/global/2018/06/19/social-media-use-continues-to-rise-in-developing-countries-but-plateaus-across-developed-ones/

15.

Seifert

Doh

Wahl

H. W.

(2017). They also do it: Internet use by older adults living in residential care facilities. Educational Gerontology, 43(9), 451–461.

16.

Shapira

Barak

Gal

(2007). Promoting older adults’ well-being through Internet training and use. Aging & Mental Health, 11(5), 477–484.

17.

Van Deursen

A. J.

Van Dijk

J. A.

(2014). The digital divide shifts to differences in usage. New Media & Society, 16(3), 507–526.

18.

Wagner

Hassanein

Head

(2010). Computer use by older adults: A multi-disciplinary review. Computers in Human Behavior, 26(5), 870–882.

19.

Werner

J. M.

Carlson

Jordan-Marsh

Clark

(2011). Predictors of computer use in community-dwelling, ethnically diverse older adults. Human Factors, 53(5), 431–447.

20.

Yoon

Jang

Vaughan

P. W.

Garcia

(2020). Older adults’ Internet use for health information: Digital divide by race/ethnicity and socioeconomic status. Journal of Applied Gerontology, 39, 105–110.

Trends in Internet Use Among Older Adults in the United States,2011–2016