Chinese Middle-Aged and Older Adults’ Internet Use and Happiness: The Mediating Roles of Loneliness and Social Engagement

Abstract

Little research has examined the effects of internet use in promoting well-being among middle-aged and older people using large-scale samples in developing countries. Using a subsample of a national survey with 4,083 adults aged 50 years and above, we explored this topic in China. Internet users were found to be significantly younger, more educated, with a higher SES, and more likely to live in urban areas. Structural equation modeling showed that internet use was indirectly associated with higher levels of happiness, and this association was mediated by less loneliness and more volunteering. In the multi-group analysis, age, gender, and household registration (urban vs. rural) moderated path coefficients but did not impact the general pattern. This study suggests promising benefits of internet use for Chinese middle-aged and older adults. In addition, the digital divide between our urban and rural subsamples calls for interventions to promote internet use in rural areas.

Keywords

Internet use happiness loneliness volunteering digital divide middle-aged and older

Introduction

With the advancement of information and communication technologies (ICTs), online engagement has become an important part of daily life. Recent data indicates that more middle-aged and older people have started to use the internet. In 2018, more than 103 million Chinese netizens were aged 50 years and above, and the number has increased by 19% from the previous year (China Internet Network Information Center [CNNIC], 2019). Given the increasing share of ICTs in everyday life, understanding the role of internet use for the middle-aged and older population is essential. Gerontological research has paid greater attention to the benefits of the internet on older populations related to well-being and has accumulated numerous findings in many countries. Most studies have been conducted in developed countries and examined older people aged 65 years and above. However, limited research has been done in developing countries or focused on the middle-aged and older population. This study aims to fill the gap by exploring the effect of internet use in promoting well-being for Chinese middle-aged and older adults.

Compared with developed countries, developing countries have a much lower internet penetration rate for middle-aged and older people. In the United States, 88% of people aged 50 to 64 years and 73% of people aged 65 years and above use the internet (Pew Research Center, 2019). While in China, about 24% of people aged 50 to 59 years and 21% of those aged 60 years and above use the internet (CNNIC, 2019; National Bureau of Statistics of China, 2019). Therefore, middle-aged and older people can be a more representative sample of “older” internet users in developing countries. Besides, developing countries have a larger rural population and lower retirement age (around 60 years). Thus, many middle-aged and older adults in developing counties who are retired or live in rural areas could experience isolation and loneliness. And, their living status may resemble those of older age in developed countries, for example, narrowed living circle, much time at home, and less economic resource. Likewise, internet use could have similar effects. Therefore, in the following, we review the literature on ICT use of middle-aged and older adults, and related evidence in China, as well as draw reference from literature examining the mechanism of internet use and older people’s well-being.

Internet Use Among Middle-Aged and Older Adults

There is rather scarce evidence concerning internet use among middle-aged and older adults. Like older adults, this age group also encounters the physiological barriers such as bad memory in learning internet skills (Lin et al., 2013) and shows lower levels of internet self-efficacy (Chu, 2010). Internet use can benefit middle-aged and older adults both cognitively and socially. One neuroimage study demonstrated that among middle-aged and older adults, prior experience with internet searching could affect the brain’s responsiveness in decision-making and complex reasoning (Small et al., 2009). Hogeboom et al. (2010) reported that internet use is positively associated with frequency of contact with friends and family, and attendance of organizational meetings for U.S. adults above 50 years.

General findings suggest the positive effects of internet use for middle-aged and older adults in China. However, most of these studies used small and convenience samples and with urban populations only. Xie (2007, 2008a, 2008b) has conducted a series of qualitative studies by interviewing participants aged 50 years and above from an internet training organization in Shanghai. Participants indicated that the training gave them a new purpose in life, let them meet new “net friends,” and encouraged them volunteering. Thus, the author argued that internet training and use could enrich post-retirement life, help build social relationships, and promote civic engagement, which then resulted in better psychological well-being. Another study interviewed 17 older people in urban China (in their 50s to 70s), and the findings suggested that going online helped participants gain quality friendships and companionship of peers (Y. Sun et al., 2014). Chiu (2019) examined internet use of this age group in Taiwan and found that active internet use was associated with increased social engagement. Several quantitative studies have also shown various benefits, such as better cognitive performance (S. Sun et al., 2016), and greater psychological well-being, especially for those living without children (Wang, 2018) or those frail older adults (Fang et al., 2018).

Internet Use, Loneliness, and Social Engagement

Many middle-aged and older people experience narrowing social circles, changing social roles, and difficulties in maintaining social relationships. These changes make them more vulnerable to loneliness and social isolation. Internet use can enable older adults to communicate with family members and friends and to complement their face-to-face interactions (Sum, Mathews, Hughes, & Campbell, 2008). A meta-analysis of intervention studies between 2001 and 2012 and found a statistically significant effect for decreased loneliness, suggesting that computer training and intervention programs might be a useful approach to assist older adults to manage their loneliness (Choi et al., 2012). Using a sample of people aged 50+ years from the 2012 Health and Retirement Study, Chopik (2016) also found that using social technologies was associated with better subjective well-being and other mental health benefits, and the associations were mediated through reduced loneliness.

Moreover, research suggests that different types of internet use could lead to different outcomes. Sum, Mathews, Pourghasem, & Hughes (2008) examined Australian internet users aged 55 years and older and found that general online communication was associated with reduced social loneliness while communicating with unknown people was associated with a higher level of emotional loneliness. One recent longitudinal study (Szabo et al., 2019) also suggested that using the internet for social purposes can decrease loneliness, thus indirectly improving well-being, whereas informational and instrumental uses were not related to loneliness.

In addition to reducing loneliness, many studies have suggested that internet use can affect older adults’ well-being by increasing their social engagement. Social engagement can be broadly described as “having connections with people and the community,” which has been widely linked to the health and well-being of older individuals (Adams et al., 2011). Engaging in social activities such as volunteering can increase a person’s social network, power, prestige, resources, and emotional gratification, all of which promote well-being (Morrow-Howell et al., 2003). Research has demonstrated that internet use can provide opportunities for and motivate older people to participate in social and community activities. For example, Ihm and Hsieh (2015) found among U.S. older adults aged 60 years and above that ICT use was associated with more social engagement.

Digital Divide and Digital Inequality

In addition, diversity within the age group should be considered. The digital divide, referring to gaps between people who have access and not, and digital inequality, emphasizing the inequality among those already have formal access to the internet, have attracted much attention (DiMaggio & Hargittai, 2001). Findings have indicated that adults who were younger, married, White, of higher educational levels, living in urban areas, and with higher income were more likely to access the internet (Berner et al., 2015; Gell et al., 2015). Gilleard and Higgs (2008) argued that with internet usage among older adults keep raising, the gap would be soon closed. However, Friemel (2016) illustrated that the process of closing access gaps is rather slow as many older adults may stop using the internet due to accumulated physical limitations. A study examined a national sample of American adults above 50 years and found age and employment interactively influence internet access (Yu et al., 2016a). For example, the odds of internet access among those unemployed and homemakers are consistently and increasingly lower than those employed with age.

Furthermore, internet technology may benefit differently among subgroups of older adults. Whether the internet benefits advantaged groups or disadvantaged groups more is still under debate. For example, research on digital inequality has illustrated that among U.S. older adults, those of higher socioeconomic status (SES) and educational level have better internet skills and are more likely to use the internet in beneficial and capital-enhancing ways (Hargittai & Dobransky, 2017). However, Gell and colleagues (2015) found that technology use significantly benefits U.S. older adults who have limitations in memory, physical capacity, or vision.

Present Study

In this study, we investigated the relationship between internet use and well-being in terms of happiness. As suggested by previous research, more internet use is associated with higher levels of well-being through its correlation with less loneliness and more social engagement among the older population, this mechanism could also be applied for middle-aged and older people in China. We, therefore, investigate whether loneliness and social engagement could mediate the relationship between internet use and happiness among Chinese middle-aged and older people. Four hypotheses were proposed accordingly:

Hypothesis 1 (H1): Internet use is associated with reduced perceived loneliness.

Hypothesis 2 (H2): Internet use is associated with more volunteering.

Hypothesis 3 (H3): Internet use has indirect effects on happiness through reduced loneliness.

Hypothesis 4 (H4): Internet use has indirect effects on happiness through more volunteering.

Furthermore, we investigated the digital divide and inequality by examining the moderating effects of demographic variables with a focus on urban and rural differences, age, and gender differences. Concerning internet use, we centered on social and communicative use. This is because social internet use is especially related to higher levels of well-being, greater social support, and connectedness (Yu et al., 2016b).

Therefore, this research intends to contribute to the status quo in three ways. First, up to now, there has been much less quantitative evidence collected in developing countries to support the effects of internet use in developed countries. This study fills the gap by examining the effects of internet use among the middle-aged and older population in China with a nationally representative data set. Second, with a large sample of the general population, we are also able to examine group differences between gender, between age groups, and between rural and urban, all of which are often closely linked to the digital divide and digital inequality. Third, structural equation modeling (SEM) was employed so that the indirect effect of internet use could be estimated. And, in our models, internet use was captured by different online activities, which can further explain the mechanism of how internet use affects middle-aged and older people’s well-being in China.

Method

Data

The Chinese Social Survey (CSS) is a national representative survey project conducted by the Chinese Academy of Social Science since 2005. The survey used China’s 2000 and 2010 censuses as the sampling frame and employed a multi-stage stratified sampling procedure, which covers 7,000 to 10,000 families and with respondents aged from 18 to 69 years old in 31 provinces and municipalities in China. We used 2015 CSS and excluded responses from residents who were younger than 50 years old, resulting in a final sample of 4,083 respondents. We chose 50 as the age limit mainly because it is the retirement age for most female workers (60 for males) in China.¹ In our sample, 784 participants were employed (19.2%), 1,508 were retired (36.9%), and 1,791 were agriculture workers (43.8%).

Measurement

Internet use

Four variables measuring internet use were constructed. In CSS 2015, participants were asked “Do you have internet skills (including using mobile phones to access the internet)?” We created the variable internet skills (1 = No, 2 = Yes) based on this question. Participants who answered “Yes” were then asked about their frequency of internet use during the last 12 months for “Using Weibo/WeChat to chat and to make friends” (chat), “commenting in forums, on blogs, or Weibo/WeChat” (comment), and “Organizing or joining offline activities (via the internet)” (offline activities). Participants rated the frequency of these activities from “almost every day,” “3 to 5 times a week,” “1 to 2 times a week,” “at least 1 time a month,” “a few times a year,” and “never.” We recoded them from 0 to 5 with higher scores indicating higher frequency, and participants who answered “no” for internet skills were also coded as 0. Previous studies have often operationalized internet use as a dichotomous variable (e.g., Gilleard & Higgs, 2008), while others have argued that the frequency of different types of use can capture finer details (e.g., Szabo et al., 2019). In this study, we used both types of measures.

Loneliness and happiness

Participants were asked to indicate from 1 (“strongly disagree”) to 5 (“strongly agree”) how much they agree with the statements “I often feel very lonely and helpless” and “Overall, I am happy.” These two questions were used to measure the respondent’s perceived loneliness and level of happiness, respectively, and higher scores indicated greater loneliness or happiness.

Volunteering

Participants were asked whether they had participated in volunteer activities or public welfare activities. We created a variable volunteer (1 = No, 2 = Yes), in which participating in either volunteer or public welfare activities was coded as “yes.”

Control variables

We included sociodemographic variables: age (in years), gender (1 = Male, 2 = Female), education (in years), marital status (1 = Never married/widowed/divorced, 2 = Married/remarried/cohabitating), family members (number of family members who currently live with the respondent), household registration (1 = rural, 2 = urban), and self-evaluated SES. SES was measured by the question “Which level of socioeconomic status in your area best describes you? (1 = bottom level, 5 = upper level).”

Analysis

We used SEM to conduct the analysis. To improve the models, variables that had no significant effect or bring no improvement were removed in the final trimmed models. Two mediation paths proposed were calculated using the product of involving paths and then tested by bootstrapping with 1,000 replications. The analysis was performed in R using the “lavaan” (latent variable analysis) package (Rosseel, 2012). Two endogenous variables (internet use and volunteering) were declared as ordered, and diagonally weighted least squares (DWLS) was used as the estimator.

Results

Table 1 reports the descriptive statistics of all variables. Men rather than women, χ²(1) = 33.10, p < .001, and married people rather than single individuals, χ²(1) = 6.31, p = .012, were more likely to have internet skills, so as people who lived in urban areas, χ²(1) = 546.75, p < .001. Internet users were significantly younger, t(767) = 11.53, p < .001, more educated, t(869.86) = −34.69, p < .001, and reported a higher SES, t(741.64) = −9.81, p < .001, but had fewer family members living with them, t(834.85) = −4.69, p < .001.

Table 1.

Descriptive Statistics by Internet Use.

Variables	Internet nonusers (N = 3528)		Internet users (N = 555)
Variables	M/%	SD	M/%	SD
Age	59.34	5.80	56.46	5.43
Male	46.23		59.46%
Education (year)	6.34	4.02	11.52	3.13
Urban	20.58%		68.11%
Family members	2.08	1.73	1.77	1.42
Married	89.20%		92.79%
SES	2.09	0.91	2.49	0.90
Loneliness	2.89	1.20	2.33	1.11
Volunteering	14.60%		34.77%
Happiness	3.47	1.06	3.81	0.98
Chat			2.97	2.13
Comment			0.97	1.66
Offline activities			0.35	0.93

Note. All variables presented significant differences between users and nonusers (p < .05). SES = socioeconomic status.

Age was negatively correlated with chat, r(553) = −.14, p = .001, but was positively correlated with taking part in offline activities, r(553) = .09, p = .04. In addition, women tended to chat more on the internet than their male counterparts, t(515.93) = −4.92, p < .001, and single people tended to comment more than married ones, t(42.62) = 2.15, p = .04.

Moreover, younger respondents, t(1,021.4) = 5.22, p < .001; men, χ²(1) = 24.20, p < .001; more educated, t(1,067.6) = −13.24, p < .001; rural residents, χ²(1) = 28.21, p < .001; and those who reported higher SES, t(1,022.1) = −4.85, p < .001, were more likely to volunteer. Perceived loneliness was negatively associated with age, r(4,081) = −.04, p = .01; education, r(4,081) = −.17, p < .001; number of family members, r(4,081) = −.04, p = .01; and SES, r(4,081) = −.30, p < .001. Older adults who were single, t(507.79) = 8.38, p < .001, or lived in rural areas, t(2,018.5) = 10.06, p < .001, reported feeling lonelier. Happiness was positively related to age, r(4,081) = .05, p < .001; education, r(4,081) = .13, p < .001; and SES, r(4,081) = .32, p < .001. Those who lived in urban areas, t(2,111.8) = −7.62, p < .001, or had a spouse, t(501.5) = −5.55, p < .001, also felt happier.

Four models were constructed using different measures of internet use. Internet skills measure the general adoption of the internet, often related to digital divide (access), while the other three measures differential use of the internet, examining digital inequality (post-access usage) and illustrating nuances within social use. Chat shows interpersonal online connections privately or in a small group. Comment indicates the public expression of opinions online, often through an online community. Offline activities reflect using the internet as a tool to coordinate social gatherings in real life. These three variables are positively correlated with each other (r_chat-comment =.572, p < .001; r_{chat-offline activities} =.406, p < .001; r_{comment-offline activities} =.439, p < .001). Still, previous research has indicated that the way in which people use the internet may influence the effect of ICT use such as differing outcomes in well-being (Szabo et al., 2019). We suggest that these four variables can capture internet use in different perspectives and thus may impact subjective well-being through loneliness and volunteering differently. For example, chatting and commenting online might be related more to reduced loneliness while offline activities are likely to be connected with more volunteering.

Model 1: Internet Skills

In Model 1 (Figure 1), internet skills was related to less loneliness (β = −.19, p < .001) and higher chance of participating in voluntary work (β = .25, p < .001). Loneliness was negatively associated with happiness (β = −.31, p < .001), while engagement in volunteering was positively associated with happiness (β = .08, p < .001). Internet skills by themselves was not related to happiness, although possessing internet skills did have a positive indirect effect on happiness through loneliness (β = .06, p < .001, 95% confidence interval [CI] = [.030, .057]) and volunteering (β = .02, p = .005, 95% CI = [.005, .024]).

Figure 1.

Model 1: Coefficients reported in the path diagram are standardized.

Model 2: Chat

Model 2 (Figure 2) showed that online chatting had similar effects as general internet skills. Chatting was negatively associated with self-reported loneliness (β = −.06, p < .001), but was positively related to volunteering (β = .09, p < .001). The effect of chatting on happiness was also fully mediated by loneliness (β = .02, p < .001, 95% CI = [.007, .023]) and volunteering (β = .01, p = .009, 95% CI = [.002, .010]).

Figure 2.

Model 2: Coefficients reported in the path diagram are standardized.

Model 3: Comment

Comment in Model 3 (Figure 3) was not significantly associated with volunteering, but it was negatively associated with loneliness (β = −.03, p = .04). Also, comment was positively associated with happiness, and this relationship was mediated by loneliness (β = .01, p = .02, 95% CI = [.003, .030]).

Figure 3.

Model 3: Coefficients reported in the path diagram are standardized.

Model 4: Offline Activities

In Model 4 (Figure 4), participating in offline activities was not associated with loneliness, but was positively related to volunteering (β = .06, p = .001). Furthermore, joining offline activities had an indirect effect on happiness, mainly through volunteering (β = .01, p = .03, 95% CI = [.003, .030]).

Figure 4.

Model 4: Coefficients reported in the path diagram are standardized.

We summarize the direct and indirect effects of internet use for the four models in Table 2.

Table 2.

Direct and Indirect Effects of the Internet Use.

Effect	Full-sample analysis (N = 4,083)
Effect	Model 1: Internet Skills	Model 2: Chat	Model 3: Comment	Model 4: Offline Activities
Internet^a → Loneliness	−.19 ***	−.06 ***	−.03 *
Internet → Volunteering^b	.25 ***	.09 ***		.06 **
Loneliness → Happiness	−.31 ***	−.31 ***	−.31 ***	−.31 ***
Volunteering → Happiness	.08 ***	.08 ***	.08 ***	.09 ***
Internet → Happiness (via Loneliness)	.06 ***	.02 ***	.01 *
Internet → Happiness (via Volunteering)	.02 **	.01 **		.01 *

Note. Standardized regression coefficients (β).

“Internet” stands for Internet Skills in Model 1, Chat in Model 2, Comment in Model 3, and Offline Activities in Model 4.

Volunteering: 1 = No, 2 = Yes; Internet Skills: 1 = No, 2 = Yes.

p < .05. **p < .01. ***p < .001.

Multigroup Analysis

To test the effects of the digital divide, we performed a multi-group analysis. The full sample was split into groups based on household registration (rural: N = 2,979; urban: N = 1,104), age (<60: N = 2,093; ≥60: N = 1,990), and gender (male: N =1,961; female: N = 2,122).

For each set of groups and each model described above, a free model, where all coefficients were allowed to vary freely, and a fully constrained model, in which path estimates were fixed to be the same between groups, were constructed. Next, a chi-square test was used to see if there was a significant difference between the two models. A significant difference in a chi-square test indicates that at least one path coefficient across these groups is not equal. To examine the difference between path coefficients, Z score, along with the corresponding p-value, was calculated.

First, we looked at the differences between urban and rural. There was a significant group difference for Model 2, χ²(21) = 55.70, p < .001, but not for the other models. Examination of paths indicated that the effects of age and education on chat were significantly different, as were the relationships between marital status and loneliness. For those living in rural areas, younger age and higher education seemed to have a weaker correlation with more frequent chatting online than for their urban-living counterparts (age: β_rural = −.13, β_urban = −.16, p = .001; education: β_rural = .16, β_urban = .32, p < .001). However, being unmarried was positively associated with loneliness and has a much stronger association with higher levels of loneliness in rural areas than in urban areas (β_rural = −.13, β_urban = −.05, p = .003).

Next, comparing the two age groups also indicated a significant group difference in Model 2, χ²(23) = 47.76, p= .002. To be specific, people living in urban areas had a higher probability to chat for those who were under 60 years than for those who were 60 years or older (β <60 = .24, β ≥60 = .19, p < .001). Similarly, higher SES and education level were also associated more significantly with online chatting for the younger group than for the older group (SES: β <60 = .09, β ≥60 = .05, p = .01; education: β <60 = .24, β ≥60 = .20, p < .001). Living in urban areas had a stronger correlation with less loneliness for the older people rather than the younger ones (β <60 = −.06, β ≥60 = −.14, p = .01).

Finally, we examined the moderating role of gender. A significant difference was found in Model 3, χ²(22) = 39.24, p = .01. Higher level of education was more strongly associated with commenting online for men (β_female = .10, β_male = .16, p = .02). However, the negative relationship between education and loneliness was stronger among women (β_female = −.14, β_male = −.03, p = .008). In rural areas, men were likely to feel lonelier than women (β_female = −.04, β_male = −.15, p = .001), and loneliness was more negatively associated with happiness (β_female = −.27, β_male = −.35, p = .002).

Discussion

In this research, we examined how internet use can affect Chinese middle-aged and older adults’ well-being with a nationally representative sample. Four hypotheses were tested and supported by four models with different measures of internet use in terms of general internet skills and frequency of online activities. To summarize, the results imply that internet use is associated with better well-being of Chinese middle-aged and older and the relationship was mediated by lower levels of perceived loneliness and more volunteer activities.

Our findings are consistent with previous studies that illustrate the positive effects of internet use in reducing loneliness. More online chatting was associated with less loneliness, as the internet allows people to contact their family members and maintain their social networks more conveniently. Similarly, commenting online is also beneficial, which allows middle-aged and older adults to present opinions among group members or to the public. Overall, internet skills can keep them more connected with others in the digital era.

Regarding volunteering, our findings suggested that the social use of the internet was associated with more social engagement, which contradicts the findings of Ihm and Hsieh (2015). In their findings, they reported that only instrumental ICT use, rather than social use, predicted social engagement. However, we found that Chinese middle-aged and older adults who use the internet to communicate and organize offline activities were more likely to volunteer. Social use of the internet can extend their social networks, increase social capital, and enhance their sense of community, which in turn increases their participation in volunteer work. For example, in China, some middle-aged adults who had previously attended computer training classes became volunteer instructors assisting their peers to navigate the web and joined internet-promoting projects (Xie, 2008a).

Our models also indicated that internet use was indirectly associated with higher levels of happiness, and the relationship was mediated by less loneliness and more volunteering, which is consistent with the findings of Szabo and colleagues (2019). This converges with other evidence supporting the multi-level links between internet use and the mental health of older adults, namely that internet use can enhance interpersonal interaction and empower social inclusion at the society level (Forsman & Nordmyr, 2017). The present study suggests that the internet can not only improve the personal lives of middle-aged and older adults by making communication easier but also reconnect them to society and enrich their public life. The internet allows them to express their needs and opinions, which could help them communicate better with the government, facilitate understanding between generations, as well as raise the age group’s voice in the digital age. However, our findings also showed internet use of Chinese middle-aged and older adults was largely confined to online chatting. Substantially fewer respondents utilized other functions of the internet, suggesting that helping older adults explore different activities online may bring more benefits in the future.

Regarding diversity within the age group, our findings suggest that there is not a significant gender gap in internet use in China. This is consistent with previous results in the U.S. indicating that gender does not contribute to disparities in ICT access (e.g., Friemel, 2016). Our models also corroborate that gender does not moderate the effect of internet use on well-being. Research has suggested that older adults are particularly vulnerable to digital inequality because they are often the last group to adopt the internet (Choi, 2011). In our sample, only 13.6% who were aged 50 years or older used the internet. The much smaller size of internet users might account for the moderating effect of age group, which showed that SES, education, and household registration were more significantly associated with chatting frequency among the younger group.

We are more interested in urban and rural differences since less research examined this question. For one thing, a digital divide does exist between urban and rural. In 2018, the internet penetration rate in Chinese rural areas (38.4%) is much lower than that in urban areas (74.6%) (CNNIC, 2019). We also observed that middle-aged and older adults who live in rural areas were less likely to use the internet and use it less frequently for different social activities. However, the digital divide within the urban sample is more salient than in rural areas since younger age and better education have stronger effects in online chatting. For another, there is no direct evidence supporting clear unequal benefits of the internet between people in urban and rural areas. Results indicated that the internet seemed to benefit them similarly. Nevertheless, middle-aged and older adults in rural areas were more fragile in many ways. Age-related declines in network size, social engagement, and social support have been found to be particularly pronounced in rural districts (Huxhold & Fiori, 2018). Our analysis also supported this that in rural areas, people felt more isolated. Compared with urban residents, men and those who were not married seemed to feel lonelier, and loneliness was associated with a greater reduction in happiness for men. In this regard, we suggest that middle-aged and older adults living in rural areas are in a greater need for social support than those in urban areas. These differences indicate that more effort should be made to promote the internet in rural areas and that the internet can be very effective in successfully coping with aging for rural residents.

Indeed, there is a growing need for middle-aged and older adults in rural areas to learn to use the internet. Volunteer activities where students teach farmers to use the internet and smartphones has been welcomed in many places, and applications such as mobile karaoke have become one of the most popular activities in villages (Chinese Academy of Social Science & Tencent Research Institute, 2018). Thus, promoting internet use to middle-aged and older adults in rural areas will be highly effective in providing social support, enriching daily life, and alleviating loneliness. Future research and interventions on reducing loneliness among rural populations are advised as well.

Limitations

In the end, several limitations must be noted. First, we used cross-sectional data, so cannot offer solid causal conclusions, and thus, our findings should be interpreted carefully. Future research based on longitudinal data or randomized controlled trials is still needed. Second, since we used previously collected survey data, methodological issues may arise as we applied a relatively constrained measure of internet use. For example, we could not know from the data whether the respondent actually accessed the internet with assistance from family members or friends. Last but not least, the proportion of internet users among Chinese older adults in 2015 was still small and this study focuses on relatively younger older adults. With a larger and older user base in the future, studies could examine how the internet influences older adults differently, such as differences between the young–old and the old–old or between those who are still working and retired.

Footnotes

Author Contributions

Y.X. planned the study, supervised the data analysis, wrote, and revised the paper. Y.H. performed statistical analysis and wrote the paper.

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the National Natural Science Foundation of China (grant no. 71902113).

ORCID iDs

Yi Xu

Yijie Huang

Notes

References

Adams

K. B.

Leibbrandt

Moon

(2011). A critical review of the literature on social and leisure activity and wellbeing in later life. Ageing & Society, 31(4), 683–712.

Berner

Rennemark

Jogreus

Anderberg

Sköldunger

Wahlberg

Elmståhl

Berglund

(2015). Factors influencing Internet usage in older adults (65 years and above) living in rural and urban Sweden. Health Informatics Journal, 21(3), 237–249.

China Internet Network Information Center. (2019). The 43rd China statistical report on Internet development. http://cnnic.com.cn/IDR/ReportDownloads/

Chinese Academy of Social Science & Tencent Research Institute. (2018). The online lives of the middle aged and the old [Research Report]. https://tengyun.tencent.com/storage/source180712/files/中老年互联网生活研究报告.pdf

Chiu

C. J.

(2019). Relationship between internet behaviors and social engagement in middle-aged and older adults in Taiwan. International Journal of Environmental Research and Public Health, 16(3), Article 416.

Choi

Kong

Jung

(2012). Computer and internet interventions for loneliness and depression in older adults: A meta-analysis. Healthcare Informatics Research, 18(3), 191–198.

Choi

(2011). Relationship between health service use and health information technology use among older adults: Analysis of the US National Health Interview Survey. Journal of Medical Internet Research, 13, Article e33.

Chopik

W. J.

(2016). The benefits of social technology use among older adults are mediated by reduced loneliness. Cyberpsychology, Behavior, and Social Networking, 19(9), 551-556.

Chu

R. J. C.

(2010). How family support and Internet self-efficacy influence the effects of e-learning among higher aged adults—Analyses of gender and age differences. Computers & Education, 55(1), 255–264.

10.

DiMaggio

Hargittai

(2001). From the “digital divide” to ‘digital inequality’: Studying Internet use as penetration increases. Princeton: Center for Arts and Cultural Policy Studies, Woodrow Wilson School, Princeton University, 4(1), 4-2.

11.

Fang

Chau

A. K.

Wong

Fung

H. H.

Woo

(2018). Information and communicative technology use enhances psychological well-being of older adults: The roles of age, social connectedness, and frailty status. Aging & Mental Health, 22(11), 1516–1524.

12.

Forsman

A. K.

Nordmyr

(2017). Psychosocial links between Internet use and mental health in later life: A systematic review of quantitative and qualitative evidence. Journal of Applied Gerontology, 36(12), 1471–1518.

13.

Friemel

T. N.

(2016). The digital divide has grown old: Determinants of a digital divide among seniors. New Media & Society, 18(2), 313–331.

14.

Gell

N. M.

Rosenberg

D. E.

Demiris

LaCroix

A. Z.

Patel

K. V.

(2015). Patterns of technology use among older adults with and without disabilities. The Gerontologist, 55(3), 412–421.

15.

Gilleard

Higgs

(2008). Internet use and the digital divide in the English longitudinal study of ageing. European Journal of Ageing, 5(3), Article 233.

16.

Hargittai

Dobransky

(2017). Old dogs, new clicks: Digital inequality in skills and uses among older adults. Canadian Journal of Communication, 42(2), 195–222.

17.

Hogeboom

D. L.

McDermott

R. J.

Perrin

K. M.

Osman

Bell-Ellison

B. A.

(2010). Internet use and social networking among middle aged and older adults. Educational Gerontology, 36(2), 93–111.

18.

Huxhold

Fiori

K. L.

(2018). Do demographic changes jeopardize social integration among aging adults living in rural regions? The Journals of Gerontology, Series B, 74, 954–963.

19.

Ihm

Hsieh

Y. P.

(2015). The implications of information and communication technology use for the social well-being of older adults. Information, Communication & Society, 18(10), 1123–1138.

20.

Lin

Y. C.

Liang

J. C.

Yang

C. J.

Tsai

C. C.

(2013). Exploring middle-aged and older adults’ sources of Internet self-efficacy: A case study. Computers in Human Behavior, 29(6), 2733–2743.

21.

Morrow-Howell

Hinterlong

Rozario

P. A.

Tang

(2003). Effects of volunteering on the well-being of older adults. The Journals of Gerontology Series B, 58(3), S137–S145.

22.

National Bureau of Statistics of China. (2019). China statistical yearbook 2019. http://www.stats.gov.cn/tjsj/ndsj/2019/indexeh.htm

23.

Pew Research Center. (2019). Internet/broadband fact sheet. https://www.pewinternet.org/fact-sheet/internet-broadband/

24.

Rosseel

(2012). lavaan: An R package for structural equation modeling. Journal of Statistical Software, 48(2), 1–36.

25.

Small

G. W.

Moody

T. D.

Siddarth

Bookheimer

S. Y.

(2009). Your brain on Google: Patterns of cerebral activation during internet searching. The American Journal of Geriatric Psychiatry, 17(2), 116–126.

26.

Sum

Mathews

M. R.

Pourghasem

Hughes

(2008). Internet technology and social capital: How the internet affects seniors’ social capital and wellbeing. Journal of Computer-mediated Communication, 14(1), 202–220.

27.

Sum

Mathews

R. M.

Hughes

Campbell

(2008). Internet use and loneliness in older adults. Cyberpsychology & Behavior, 11(2), 208–211.

28.

Sun

Zhang

Fan

(2016). Media use, cognitive performance, and life satisfaction of the Chinese elderly. Health Communication, 31(10), 1223–1234.

29.

Sun

Ding

Lindtner

(2014). Being senior and ICT: A study of seniors using ICT in China [Conference session]. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp. 3933–3942.

30.

Szabo

Allen

Stephens

Alpass

(2019). Longitudinal analysis of the relationship between purposes of internet use and well-being among older adults. The Gerontologist, 59(1), 58–68.

31.

Wang

(2018). The impact of internet on the social wellbeing of the elderly in urban China. Innovation in Aging, 2(Suppl 1), Article 738.

32.

Xie

(2007). Older Chinese, the internet, and well-being. Care Management Journals, 8(1), 33–38.

33.

Xie

(2008a). Civic engagement among older Chinese internet users. Journal of Applied Gerontology, 27(4), 424–445.

34.

Xie

(2008b). The mutual shaping of online and offline social relationships. Information Research, 13(3), Article 350.

35.

R. P.

Ellison

N. B.

McCammon

R. J.

Langa

K. M.

(2016a). Mapping the two levels of digital divide: Internet access and social network site adoption among older adults in the USA. Information, Communication & Society, 19(10), 1445–1464.

36.

R. P.

McCammon

R. J.

Ellison

N. B.

Langa

K. M.

(2016b). The relationships that matter: Social network site use and social wellbeing among older adults in the United States of America. Ageing & Society, 36(9), 1826–1852.