The DVC has changed the life style, working state and behavior mode of rural residents in an unprecedented way, and has a profound impact on the happiness of rural residents.

Firstly, the DVC broadens the sources of rural residents’ information access. The information gap theory holds that unequal access to information will lead to the aggravation of social inequality (Matrosov et al., 2013). The DVC aims to reduce the information gap between urban and rural areas, enabling rural residents to also enjoy the convenience brought by digitalization, such as e-government and e-commerce, thereby reducing social inequality and enhancing the happiness of rural residents (Malecki, 2003). Relying on digital technologies such as big data and cloud computing, it is possible to effectively alleviate the information asymmetry rural residents (Kong et al., 2022). The big data technology has the function of efficiently capturing and integrating information (Liu and Tian, 2022). Meanwhile, the Internet platform can realize the multi-dimensional presentation and rapid sharing of information (Liu et al., 2021). According to the theory of information visualization, when information that is difficult to be directly displayed is transformed into forms such as graphics and videos for presentation, it can greatly enhance the individual’s reception degree of information and expand the sources of information (Munzner, 2014). Rural residents through using of the internet enrich their knowledge and vision, absorb fresh ideas, so as to improve the happiness of them (Zhang et al., 2024). Ma and Le (2019) shows that digital development creates new activities such as online leisure and entertainment, online shopping, which improve the subjective welfare level of rural residents.

Secondly, the DVC improve rural residents’ skills. The technology acceptance model explains how users accept and utilize a new technology (Davis, 1989). In DVC, as the acceptance of new technologies by rural residents increases, they can better utilize these technologies to enhance production efficiency (Zhao et al., 2024), obtain information, and enjoy services, thereby improving the quality of life and happiness (Hong and Chang, 2020). The DVC offers various online learning opportunities through the Internet, which can enhance the earning capacity of rural residents (Yin et al., 2024). Digital tools can equip rural residents with an abundance of knowledge regarding modern agricultural techniques, such as soil management, precision farming, and crop health monitoring, thereby enhancing their agricultural skills (Bai et al., 2024). Zhou and Zhang (2021) show that using the Internet can improve work ability. Ma and Le (2019) conduct empirical tests and find that online learning can significantly enhance the happiness of rural residents.

Thirdly, DVC broadens social network relationship of rural residents. The social capital theory indicates that social networks, norms, trust and other forms of social capital have a significant impact on the welfare of social members (Cook, 2001). The DVC enhances the internal connections within rural communities and their ties with the outside world by establishing a broader information network (Sabatini and Sarracino, 2017), thereby increasing social capital and improving residents’ sense of happiness (Ma and Le, 2019). The use of digital technology facilitates the communication between rural residents (Castellacci and Vinas-Bardolet, 2019). The use of mobile Internet not only helps maintain existing strong relationships (Salanova et al., 2004), but also forms new relationships, thus strengthening rural residents’ social network relationships, enhancing their sense of self-worth, and enhancing rural residents’ happiness (Castellacci and Tveito, 2018).

Based on the above analysis, we speculate that if DVC is high, rural residents will experience increased happiness due to easy access to abundant information, rapid skill improvement, and strong social relationships. Thus, we propose the hypothesis H1: the higher the DVC, the higher rural residents’ happiness.

Economic resources are the material basis of people’s happiness. The rural residents may be increased economic resources by the DVC, further increase their happiness. On the one hand, the DVC broadens the channels for entrepreneurship, increases rural residents’ income (Li et al., 2021). With the development of digital technology, rural residents can use the Internet platform to sell agricultural products abroad, which improves the value of agricultural products (Omrani and Martin, 2014). They can also use e-commerce, live video and other ways to start a new business, changing the traditional means of survival based on agriculture (Pénard et al., 2013). On the other hand, the DVC creates a variety of new job positions, and further promoted rural residents’ income (Tu and Sui, 2011). In the past, rural residents, limited by remote geographical location, inconvenient transportation and other factors, often rely on farming for income. The income obtained from planting food has great uncertainty (Sabatini and Sarracino, 2017). On the contrary, nowadays, with the application and development of digital technologies, rural residents can use digital technologies such as smart agriculture for planting agriculture, reducing the failure rate of planting agriculture (Van Gaasbeck, 2008). Furthermore, digital technology sinks to rural areas and creates a lot of jobs (Calvet et al., 2009). For example, rural residents need a large number of anchors, operation personnel, security personnel, transportation personnel and after-sales personnel to sell agricultural products through live broadcasting platforms (Jin and Xing, 2024). All of these have greatly increased the employment rate in rural areas, achieved employment at the doorstep, obtained higher income, and greatly increased the happiness of rural residents (Ma and Le, 2019). Based on the above discussion, we propose the research hypothesis H2: the DVC can obtain happiness through rural residents’ income.

Loneliness is a reflection of a closed mind, which is the feeling of isolation and exclusion from the outside world (Kraut et al., 1998). Loneliness is a feeling that rural residents are generally ignored by the public, and this feeling will undoubtedly affect the happiness of them (Castellacci and Tveito, 2018). They feel lonely, because their children leave the countryside and live in the city (Morrish, 2021). There are generally no recreational activities and social activities for rural residents (Diener et al., 1999). The DVC has a positive influence on reducing the loneliness of them (Nie et al., 2021). Through social media, instant messaging and other tools, DVC helps rural residents establish and maintain connections with family, friends and neighbors (Baric et al., 2018). Even if rural residents are in the countryside, they can enjoy similar social interactions with urban people, thus reducing their loneliness. Many elderly people learn to use smart phones, and sell agricultural products through the Internet (Brooks, 2015). The rural residents buy daily products, enjoy convenient remote services, which not only improves their quality of life, but also increases the interaction with the outside world (Castellacci and Vinas-Bardolet, 2019).

Based on the above analysis, we propose the research hypothesis H3: the DVC can improve rural residents’ happiness by reducing their loneliness.

The data in this paper includes two parts. The first part is the Index of Digital Rural County (IDRC). The Index is released by the Institute of New Rural Development of Peking University and Ali Research, which is used to describe the development of Chinese county-level digital countryside (New Rural Development Institute, Digital Village Project Team, Peking University, 2020). Specifically, it has four sub-dimensions, including digital infrastructure index, digital rural economy index, digital rural governance index and digital rural general index.

The second part of the data comes from the China Family Panel Studies (CFPS) funded by Peking University and the National Natural Science Foundation of China. The CFPS is maintained by the Institute of Social Science Survey of Peking University. The CFPS data is organized and implemented by the China Social Survey Center of Peking University, using multi-stage equal probability sampling, the sample covered 162 counties in 25 provinces, and the target sample size was 16,000 households. 2010 was the base period for sampling and research, and the following survey is conducted every 2 years. The data utilized in this study pertain to the years 2018 and 2020, chosen specifically as IDRC data became accessible exclusively in these years.

The data is processed as follows: (1) Urban samples are excluded of CFPS and IDRC, only rural samples are retained. (2) Only family samples and matched household head characteristics are retained for CFPS. (3) Because the digital rural Index of Beijing, Tianjin, Shanghai and Chongqing is not collected in IDRC for 2018 and 2020, the samples of these four cities are deleted. Finally, we match IDRC with CFPS by county ID. The sample data of 8,187 rural households were obtained after removing incomplete or anomalous sample.

The explained variable is rural residents’ happiness (Happiness). Referring to the existing studies, the answer of “How happy do you feel?” of the CFPS2018 and 2020 questionnaires are used to evaluate the rural residents’ happiness. Responses to this question are measured on a scale of 0–10, from “very unhappy” to “very happy” with higher scores indicating higher levels of happiness of rural residents.

The explanatory variable is DVC, which measured by Index of Digital Rural County (IDRC). The original value of IDRC ranges from 0 to 100. In addition, the empirical analysis part adopts the standardized DVC except the descriptive statistics part. The standardized method is to divide the raw data by 100.

The theoretical analysis and research hypotheses suggest that the mechanism by which DVC enhances the happiness of rural residents by augmenting their income (Income) and alleviating their sense of loneliness (Lone). In order to verify these two mechanisms, the Income is measured by the average household income of CFPS. The Loneliness is measured by the question of “I feel lonely in a week” and its answer in the CFPS database. Replies to this query are gauged on a four-point scale, ranging from “1) Almost none (less than a day)” to “4) Most days (5–7 days).” This scale delineates four distinct tiers of frequency: “1) Almost none; 2) Some days; 3) Often; 4) Most days,” with ascending scores signifying intensified degrees of loneliness. “Almost none (less than a day)” to “Most days (5–7 days).” This scale delineates four distinct tiers of frequency: “1) Almost none (less than a day); 2). Some days (1–2 days); 3). Often (3–4 days); 4. Most days (5–7 days),” with ascending scores signifying intensified degrees of loneliness.

To mitigate the impact of extraneous variables on the regression outcomes, and drawing upon extant research (Jiang, 2023), the control variables are principally categorized into two groups. One groups is to describe the rural resident characteristics, including age, gender, education level, marital status, health level, occupation, religious belief, social status, and whether there is medical insurance. Another category is to measure family characteristics, including the size of family, per income of the family (Li et al., 2021). Detailed descriptions of the variables, along with their descriptive statistics, are presented in Table 1.

Considering that the dependent variable of this study, rural residents’ happiness, is ordinal and discrete, an Ordered Probit model is employed for the estimation. To investigate the effect of DVC on rural residents’ happiness, Equation 1 is established.

Where Happiness represents the happiness of rural resident i in j county. Dig represents the level of DVC in County j. Control is a set of control variables. ε represents the random error term. This paper focuses on the β₁ coefficient. According to the above theoretical analysis and research hypothesis, we predict that β₁ is positive.

Table 2 show the results of the Ordered Probit Regression for model (1). In column (1), it does not include any control variables, the coefficient on DVC is 0.063, statistically significant and positive at the 1% level. In column (2), post inclusion of control variables, the coefficient of DVC on rural residents happiness is 0.0471, which remains significantly positive at the 1% level. This substantiates the assertion that DVC indeed elevates rural residents happiness, thus confirming hypothesis H1. Regarding the control variables, the majority exert a significant influence on rural resident happiness. Marital status, health condition, education level, social status, participation in medical insurance, and household per capita income all have a notable positive impact on rural residents’ happiness. Employment status, religious beliefs, and family size did not yield statistically significant effects on rural residents’ happiness.

Based on the practice of Ferreri Carbonell and Frijters (2004), this paper considers rural residents happiness as a continuous variable and uses Ordinary Least Squares (OLS) as a benchmark regression. If the coefficient of Dig is positive, it indicates that the DVC increases the happiness of rural residents. Column (1) of Table 3 reports the OLS regression results, and it can be seen that the coefficient is 0.0988 and still significantly positive at the 1% level, which supports the hypothesis H1.

This paper also employs the questionnaire “whether to access the internet via mobile” from CFPS as an alternative measure for the DVC. This approach is due to potential reservations in the respondents’ mindset or inconsistencies in the questionnaire response standards, which may introduce bias into the subjective appraisal of happiness. Column (2) of Table 3 shows the result. The coefficient of Dig is 0.0473, and remains positive and significantly.

To mitigate errors potentially arising during the survey, the recorded responses for the dependent variable, rural residents happiness, were reassigned. The adjustment protocol reclassifies responses indicating happiness levels from “0–5” as “0” and those from “6–10” as “1.” The Probit model was reapplied using model (1) for estimation, with the regression results presented in Column (3) of Table 3. The findings suggest that even when the measurements for rural residents happiness are altered, the impact of the DVC on rural residents happiness remain robust.

By referring to Jin et al. (2024), we further control other factors that may affect rural residents’ happiness. Especially, we add rural residents’ relative income, political status and mortgage status into the model (1). The results are shown in Column (4) of Table 3, and the coefficient of Dig is still significantly positive at 1% level. It indicates that the baseline results are reliable.

In the empirical tests, we employ the macro-level County Digital Village Index as the explanatory variable, while the explained variable, rural residents happiness, is derived from micro-level household data. Incorporating data from disparate levels into a regression model can mitigate the interference of reverse causality on the estimated results. This is predicated on the notion that a rural resident’s sentiment is unlikely to exert influence on the DVC. Therefore, the problem of omitted variable constitutes a potential endogeneity in the research. Factors such as rural residents’ social networks, cultural custom, individual acceptance of the digital technology and the level of digital technology application, could all bear upon the sense of well-being among rural residents.

Given that the dependent variable of this article is ordinal, the Conditional Mixed Process (CMP) is utilized to further address selection biases arising from endogeneity, thereby describe the causal effect of DVC on rural residents happiness. Analogous to the instrumental variable estimation technique, the CMP approach necessitates the identification of plausible instrumental variables to alleviate the endogeneity concerns associated with DVC. A valid instrumental variable must satisfy two criteria: relevance and exogeneity. That is, the instrumental variable must correlate with the endogenous explanatory variable while not directly affecting the dependent variable.

Referring to the general practice of existing literature, this study adopts the interaction between the 2017 internet penetration rate and the number of landlines per hundred people from 1984 as an instrumental variable. The underlying rationale is that DVC are intertwined with the evolution of information and communication technologies, among which landlines were one of the early widespread means of communication. The abundance of landlines in a region is closely linked to its current internet development, which, in turn, bears upon the ongoing DVC. Hence, it is reasonable to posit a positive correlation between DVC and the number of regional landlines, meeting the relevance requirement for an instrumental variable. On the other hand, the quantity of landlines in a region is unlikely to exert a direct impact on rural residents happiness, thus satisfying the condition of exogeneity.

Table 4 presents the regression outcomes, wherein the F-statistic is 24.77, surpassing the empirical benchmark of 10 and thus rejecting the null hypothesis associated with a weak instrumental variable. In the initial stage of estimation, the instrumental variable—2017 internet penetration rate and the number of landlines per hundred people in 1984—correlates significantly and positively with DVC, corroborating the positive association between the chosen instrumental variable and the endogenous variable. The secondary phase of regression reveals that DVC has an effect size of 0.0981 on rural residents happiness, retaining significance at the 1% threshold. Accordingly, the estimations derived from the CMP methodology reinforce the assertion that DVC enhances rural residents well-being, indicating that the research conclusions of this article are robust.

In order to further explore the structural effects of DVC on rural residents happiness, this part compares and analyzes the impact of DVC on rural resident happiness from four sub dimensions including rural digital infrastructure index, rural digital economy index, rural digital governance index and rural digital living index. The results are shown in Table 5 for four sub-dimensions index. According to the coefficients, the rural digital infrastructure index and rural digital economy index have a significant positive impact on the rural residents well-being. The possible explanation is that the development of digital infrastructure and the digitalization of rural economy bring a lot of convenience to people’s lives, such as transportation and logistics, financial services, health care, etc. Studies have shown that the opening of high-speed rail can affect rural poverty reduction, promote employment and income, improve consumption level, improve residents’ health status, and thus improve rural resident happiness.

The DVC may have varying impacts due to the individual characteristics of the different rural residents (Lindberg and Úden, 2010). This part selects the gender and age of the rural residents for further analysis, with the estimated results shown in column (1–2) of Table 6. The regression outcomes indicate that DVC has a more pronounced effect on the well-being of the male compared to females. The reasons that have a greater impact on the male group are as follows: First, men usually have more decision-making power in families and communities. Therefore, in the DVC, men may have a greater access to the use and benefits of new technologies. Second, men usually have more opportunities to receive formal education and technical training than women, which may lead them to be better able to take advantage of the opportunities brought by the DVC than women. Third, men’s participation in the labor market is often higher. The improvement of agricultural production efficiency and market access by digital technologies directly affects men’s careers more, thus enabling men to benefit more from the DVC.

The DVC exerts a significantly positive influence on the happiness of the middle-aged group, as opposed to the youth as shown in column (3)–(5) of Table 6. The reasons why the impact is significant on the middle-aged group but not so obvious on the young group are as follows: First, although the young group may be more familiar with and receptive to new technologies, the middle-aged group’s acceptance and use of technology are often driven by practical needs and problem-solving. The technical services provided by DVC can help the middle-aged group solve practical problems, thereby generating more significant positive effects in their lives. Second, the middle-aged group may need more diverse information acquisition channels, including agricultural technology, market information, policy guidance, etc., and DVC can precisely provide such information services. While the young group are “digital natives,” the new digital technologies bring less novelty to them and have a smaller marginal effect, so their influence is smaller.

The impact of DVC on rural residents happiness may vary due to differences in educational attainment. Based on Jiang al. (2023), this part defines the education level below high school as “low education,” and defines the education level above high school as “high education.” The estimated results are shown in column (6)–(7) of Table 6. According to the regression results, the DVC has a significant positive impact on the happiness of the low education groups.

This is because residents with low education groups have lower cultural quality, less knowledge reserve and less learning ability than those with high education groups. With the DVC, rural residents with low education groups have more channels to access digital technology, and can benefit more from it, which has a stronger role in improving their happiness. For the high-educated group who already enjoy a relatively high quality of life, the construction of digital countryside might only be one of the many conveniences they have already possessed, and thus its effect on enhancing their sense of happiness is rather limited. However, for the low-educated group, the same change might mean a huge leap from nothing to something, and thus their feelings would be more intense.

DVC can bring more job opportunities and entrepreneurship platforms, and improve the income level of rural residents through innovative agricultural models, e-commerce and other means (Whitacre, 2008). Through the application of digital technology, rural residents can participate in the broader economy, education and job market, have the opportunity to obtain more resources and opportunities (Brooks, 2015), improve their economic and social status, reduce the gap between the rich and the poor, and thus improve their happiness (Jin and Xing, 2024). To verify this hypothesis, this part undertakes some tests. The results of the test are shown in column (1)–(3) of Table 7.

The column (1) presents the results of the baseline regression, while column (2) delineates the effects of DVC on rural residents’ income. The column (3) incorporates an intermediary variable -rural residents’ income and explores its relationship with the DVC and rural residents’ happiness. It is discernible that, upon the inclusion of the intermediary variable, the impact coefficient of DVC on rural residents’ happiness diminishes slightly yet remains significant. Moreover, the effect of rural residents’ income on their well-being is positively significant at the 1% level, indicating that income partially mediates the relationship between DVC and rural residents’ happiness. This substantiates the theory that an increase in rural residents’ income is one of the conduits through which DVC affects rural residents’ well-being, thereby confirming hypothesis H2.

In rural areas, rural residents mainly engage in agriculture as their daily occupation and the work is arduous. Moreover, they are far away from urban areas. There are fewer entertainment programs in rural areas, which makes them feel more lonely and their sense of happiness may be lower (Castellacci and Tveito, 2018). However, the development of DVC has broken through the limitations of time and space, enabling people in remote rural areas to achieve online connections with just one click, meeting their needs for making friends, communicating, interacting, entertaining, relaxing, shopping, etc., greatly satisfying their spiritual needs and enhancing their sense of happiness (Hong and Chang, 2020). Therefore, this part examines whether the digital village construction can enhance the sense of happiness of rural residents by reducing their sense of loneliness (Briggeman and Whitacre, 2010).

The results are shown in column (4)–(6) of Table 7. Column (4) is the baseline regression. In column (5), rural resident loneliness is presented as the explained variable, while the DVC serves as the explanatory variable. It can be seen that the DVC reduces the loneliness of rural residents. In column (6), incorporating rural resident loneliness, the effect coefficient of DVC on rural residents’ happiness has decreased to some extent, yet it remains significant. The rural resident loneliness positively correlates with their happiness at the 1% significance level, suggesting that it plays a partial mediating role between DVC and rural residents’ well-being. This supports the notion that reduce rural residents’ loneliness may be one of the pathways through which the DVC influences rural residents’ happiness, thereby confirming H3.