Abstract
Background
In the context of a gradual increase in aging, improving the mental health of the elderly is particularly vital for coping with aging. Leveraging data from the 2020 China Family Panel Studies, this study rigorously examines the influence of short video on the mental health of the elderly.
Methods
We use a multiple linear regression model to investigate the influence of short video usage on the mental health of the elderly. To address endogeneity concerns, this study employs two-stage least squares and propensity score matching to estimate the impact of short video usage on the mental health of the elderly.
Results
The empirical analysis reveals a substantive and statistically significant enhancement in the mental health of elderly people attributable to the use of short videos. To ensure the reliability and robustness of our estimations, a comprehensive battery of robustness tests is conducted, all of which consistently support the conclusion of a positive association between short video usage and improved mental health among the elderly. Furthermore, the results of the heterogeneity analysis suggest that short videos have less of an impact on elderly males and individuals with higher levels of education. The results of the mechanism analysis indicate that the use of short videos can enhance the mental health of elderly individuals by positively impacting the intergenerational relationships between them and their children, as well as their leisure consumption habits.
Conclusions
This study can provide policy inspiration for the government to improve the mental health of the elderly and achieve active aging.
Keywords: Short video, Mental health, Intergenerational relationships, Leisure consumption
Introduction
China has the world’s largest elderly population and is undergoing the most rapid aging compared with other nations. Data from China’s seventh national census reveal that the elderly population aged 60 years and older totals 264 million, representing 18.7% of the total population1. According to the World Health Organization, China’s elderly population is projected to peak at 487 million by around 2050, constituting 34.9% of its total population2. Effectively addressing and mitigating the health risks faced by the elderly and achieving active aging have emerged as significant theoretical and practical concerns for governmental authorities and industry regulatory bodies across all administrative levels in China.
Currently, elderly individuals in China encounter challenges, such as inadequate access to healthcare resources, restricted social engagement, and age-related discrimination, all of which significantly jeopardize their mental well-being. According to the “2021 China Internet Audio-Visual Development Research Report,” as of March 2021, the average daily usage time of short videos in China has surged to 125 min3. Hence, it is evident that the use of social media platforms, such as short videos, has become a pivotal aspect of daily life for Chinese residents. Moreover, interpersonal interactions and communication are progressively transitioning toward the burgeoning realm of short video platforms. As the frequency of short video usage increases among residents, it may have a multifaceted impact on the lives of the elderly, consequently influencing their mental health.
The use of short videos may have both positive and negative effects on health. Elderly people may enhance their health literacy through short videos, facilitating disease prevention and timely medical interventions. Moreover, short video usage has been associated with decreased feelings of social isolation and loneliness among the elderly [1]. A short video platform featuring the ability to showcase one’s own life remains a significant avenue for attaining “happiness in old age” and bolstering social interactions with acquaintances. By engaging in learning and discussions about the functionality of short videos with relatives and friends, elderly people can not only acquire proficiency in utilizing short video social media platforms such as Douyu and Douyin, but also enhance social interactions, bolster social ties, augment their sense of social integration, and facilitate adaptation to the evolving social landscape. It is evident that the use of short videos may yield positive effects and enhance the mental health of the elderly.
The use of short videos by elderly people may also have adverse effects, potentially jeopardizing their mental health [2]. First, the elderly population is susceptible to misinformation that is prevalent in short videos, leading to deception. Second, the abundance of irrelevant, pseudoscientific, and misinformation within short videos may render the elderly’s health practices ineffective, erode their trust, and induce anxiety. Moreover, the novelty of short videos may encourage the elderly to spend excessive amounts of time viewing them. Additionally, certain short video social media platforms foster an immersive media environment, contributing to potential overindulgence among the elderly. Excessive exposure to short videos can readily precipitate feelings of anxiety and other negative emotions [3, 4]. Therefore, the use of short videos may have adverse effects, detrimentally affecting the mental health of the elderly.
In summary, the influence of short video usage on the mental health of the elderly remains a contentious topic. Determining how elderly people should adapt to this impact and delineating the role of governmental intervention are pivotal issues that require resolution. Drawing upon theoretical frameworks and practical exigencies, this study concentrates on the mental health of the elderly. Using data from the China Family Panel Studies (CFPS), we aim to scrutinize the effects of short video usage on the mental health of elderly people and elucidate its underlying mechanisms. This study seeks to offer novel perspectives and insights for enhancing the mental health of the elderly and fostering healthy aging.
The innovations of this study are reflected in the following aspects: First, in terms of research participants, existing literature on the “health effects of social media” predominantly centers around adolescents, with no consensus reached regarding the impact of social media use on mental health. By leveraging extensive microdata from the CFPS, this study employs mainstream econometric models to effectively discern the ramifications of short video utilization on the mental health of the elderly. This endeavor contributes to the enrichment of the literature concerning short video utilization and the mental health of elderly people, offering robust theoretical and empirical substantiation to comprehend the influence of short videos on elderly mental health. Second, this study examines the impact of short videos on the mental health of elderly individuals across different genders and educational levels, facilitating a nuanced understanding of its heterogeneous effects. Third, apart from investigating the direct impact of short video use on elderly people’s mental health, this study elucidates the underlying mechanisms by which short video use influences the mental health of the elderly. The findings offer valuable policy insights for governments to enhance elderly mental health, bolster elderly welfare, and foster healthy aging.
The remainder of this paper is structured as follows: Sect. 2 presents the data and models used to estimate the impact of short videos on the mental health of the elderly. Section 3 details the estimated results of this impact. Section 4 discusses our empirical findings. The final section concludes this paper.
Data and methodology
Data
Data used in this study originate from the 2020 CFPS organized and executed by the China Social Survey Center of Peking University. Currently, many scholars have used CFPS to conduct causal inference studies [5, 6]. Employing computer technology and a multi-stage sampling approach, the survey gathered data across various levels, including individuals, households, and communities. The CFPS research team has extensive experience in survey implementation, conducting biennial surveys since 2010. However, because of the absence of pertinent data on short video usage in the CFPS records preceding 2020, this study refrained from using CFPS data from 2010 to 2018. In 2020, the CFPS survey encompassed 30 provinces, autonomous regions, and municipalities, including notable regions, such as Sichuan, Guangdong, Hubei, Shanghai, and Beijing. The CFPS dataset comprises a broad spectrum and features comprehensive survey content, providing detailed data support for investigating the impact of short video use on elderly mental health. For data processing, we removed samples with missing values for short video usage, elderly health, and various control variables. Furthermore, as this study focuses on the elderly, individuals younger than 60 years of age were excluded from the sample. After data processing, 3,062 samples remained.
Variable definition
Dependent variable
The dependent variable is the mental health status of the elderly. Referring to previous studies [6–10], this study assesses the mental health of elderly people using the 8-item Center for Epidemiologic Studies Depression (CES-D) scale. The CES-D scale encompasses inquiries concerning various aspects of mental health among elderly people, including (1) The frequency with which they feel depressed. (2) The frequency with which they find everything to be a considerable effort. (3) The frequency of experiencing poor sleep. (4) The frequency of feeling unpleasant. (5) The frequency of feeling lonely. (6) The frequency of feeling unhappy about life. (7) The frequency of feeling sad. (8) The frequency of feeling that life is not worth living. Respondents are provided with four options based on their individual circumstances: (a) Almost never (less than one day). (b) Sometimes (1–2 days). (c) Often (3–4 days). (d) Most of the time (5–7 days). The above four options are assigned the values of 4, 3, 2, and 1. Utilizing the respondents’ responses to these eight questions, the mental health status of elderly people can be gauged. To mitigate potential estimation biases stemming from direct summation, this study employs factor analysis to assess the mental health of elderly people. Furthermore, to ensure the robustness of estimation outcomes, direct summation and principal component analysis techniques are employed during robustness testing to evaluate the mental health status of the elderly.
Explanatory variable
The core explanatory variable in this study is whether short videos are utilized. Within the CFPS questionnaire, a query concerning short videos is included: “Have you watched short videos such as Xiaohuoshan, Tiktok, Kuaishou, Weishi, Douyu, etc. in the past week?” If the elderly respondent answers “yes,” the core explanatory variable is assigned a value of 1; otherwise, 0. Therefore, this study constructs a dummy variable that reflects the usage of short videos among the elderly.
Control variable
When examining the influence of short video usage on the mental health of the elderly, the potential issue of omitted variables could introduce notable biases into the estimation outcomes. Referring to previous studies [11, 12], this study integrates an array of individual-, household-, and city-level control variables. At the individual level, control variables include the elderly individual’s age, gender, residency status, marital status, smoking habits, and Internet usage. Household-level control variables include indicators such as government subsidy receipt, household size, net assets, and income. The city-level control variables encompass the economic development status, total population, and industrial composition of the city where the elderly reside. Considering that certain provincial-level factors may influence the mental health of the elderly, this study further controls for housing prices, number of beds per capita, and old-age dependency ratio in the provinces where the elderly reside. Table 1 presents the definitions of the variables used in the empirical analysis.
Table 1.
Variable definition
| Variable | Definition |
|---|---|
| Mental health | Calculated based on the 8-item CES-D scale |
| Video | Dummy variable equals 1 if the elderly people utilize the short video, 0 otherwise |
| Age | Age for individual |
| Gender | 1 for male, 0 for female |
| Education | Education level |
| Registration | 1 for urban residency, and 0 otherwise |
| Marry | 1 for married, and 0 otherwise |
| Smoke | 1 for smoked, and 0 otherwise |
| Internet | Individuals who use the Internet are assigned a value of 1, while those who do not are assigned a value of 0. |
| Subsidy | Dummy variable equals 1 if the household receives government subsidies, 0 otherwise |
| Household size | The number of people in the family |
| Household asset | The net assets of the household |
| Household income | The income of the household |
| Economy | The economic development level of the city where the individual resides (in log). |
| Population | The total population of the city where the individual is located at the end of the year (in log). |
| Tertiary industry | The proportion of the tertiary industry output value to the GDP in the city where the individual resides. |
| Housing prices | The average housing prices in the province where the elderly reside (in log). |
| Beds | The number of beds per capita in the province where the elderly reside (in log). |
| Dependency ratio | The old-age dependency ratio in the province where the elderly person resides. |
Description
Table 2 presents the descriptive statistical analysis results for the mental health of the elderly, short video usage, and all control variables. From the descriptive statistics provided in Table 2, the effective sample size for each variable is 3062. Upon examining the mean values of each variable, it is notable that the mean value of the dependent variable, mental health, is 0.039. Additionally, this study presents descriptive statistical analysis results comparing the mental health of elderly individuals who use short videos with those who do not. Among these groups, the mean mental health level of elderly people who use short videos is 0.226, whereas the mean mental health level of those who do not use short videos is -0.001. In summary, the overall level of mental health among elderly people who use short videos appears to be higher than among those who do not use short videos. However, to elucidate the causal relationship between short video usage and the mental health of the elderly, further analysis is warranted through the establishment of econometric models in subsequent sections. Table 2 presents the means and standard deviations of key explanatory variables. Specifically, the mean short video usage is 17.7%, indicating that the proportion of elderly individuals who use short videos is low. The mean educational attainment is 5.89 years, suggesting that the average educational duration among the elderly is relatively low.
Table 2.
Descriptive statistics of variables
| Variable | Full samples | Video = 1 | Video = 0 | |||
|---|---|---|---|---|---|---|
| Mean | SD | Mean | SD | Mean | SD | |
| Mental health | 0.039 | 0.696 | 0.226 | 0.656 | -0.001 | 0.698 |
| Video | 0.177 | 0.381 | 1.000 | 0.000 | 0.000 | 0.000 |
| Age | 68.324 | 5.876 | 66.270 | 5.283 | 68.765 | 5.904 |
| Gender | 0.530 | 0.499 | 0.597 | 0.491 | 0.515 | 0.500 |
| Education | 5.890 | 4.684 | 8.677 | 4.177 | 5.292 | 4.571 |
| Registration | 0.491 | 0.500 | 0.632 | 0.483 | 0.461 | 0.499 |
| Marry | 0.834 | 0.372 | 0.872 | 0.334 | 0.826 | 0.379 |
| Smoke | 0.295 | 0.456 | 0.311 | 0.463 | 0.292 | 0.455 |
| Internet | 0.257 | 0.437 | 1.000 | 0.000 | 0.098 | 0.297 |
| Subsidy | 0.411 | 0.492 | 0.362 | 0.481 | 0.422 | 0.494 |
| Household size | 3.630 | 2.024 | 3.497 | 1.921 | 3.659 | 2.044 |
| Household asset | 0.912 | 2.203 | 1.552 | 3.491 | 0.774 | 1.782 |
| Household income | 10.743 | 1.221 | 11.187 | 1.053 | 10.648 | 1.234 |
| Economy | 10.883 | 0.641 | 10.969 | 0.656 | 10.865 | 0.637 |
| Population | 6.248 | 0.657 | 6.235 | 0.696 | 6.251 | 0.648 |
| Tertiary industry | 52.713 | 10.392 | 53.938 | 11.416 | 52.451 | 10.142 |
| Housing prices | 9.097 | 0.480 | 9.164 | 0.540 | 9.082 | 0.465 |
| Beds | 4.155 | 0.122 | 4.158 | 0.124 | 4.155 | 0.122 |
| Dependency ratio | 18.366 | 3.324 | 18.444 | 3.382 | 18.349 | 3.311 |
| Observations | 3062 | 541 | 2521 | |||
Methodology
Given that the mental health of the elderly is a continuous variable and the data used in this study are cross-sectional, we draw on the methodologies of previous studies [6, 13] and employ a multiple linear regression model to analyze the impact of short video use on the mental health of the elderly. The multiple linear regression model is formulated as follows:
 |
1 |
In Eq. (1), 
represents the dependent variable, which is mental health of the elderly. 
denotes the constant term estimated using the model. 
is the core explanatory variable indicating whether short videos are used. 
represents the estimated coefficient of the impact of short videos on the mental health of the elderly. If the estimated coefficient is significantly greater than 0, it indicates that using short videos can significantly improve the mental health of the elderly. X represents individual-, household-, city-, and provincial-level control variables such as age, gender, and educational attainment. 
denotes the estimated coefficients of the control variables. 
represents the error term.
Empirical results
Baseline results
Table 3 presents the estimated outcomes regarding the influence of short video usage on the mental health of the elderly, using multiple linear regression. The estimation results in Column (1) of Table 3, without the inclusion of any control variables such as age and gender, indicates that the estimated impact of short video usage on the mental health of elderly people is 0.023, which is significantly positive at the 1% significance level. After including age, gender, education, and other control variables related to individual and family characteristics, the estimation results in Columns (2) and (3) of Table 3 indicate that the estimated effects of short video usage on the mental health of elderly individuals are 0.016 and 0.012, respectively. Importantly, both estimates remain significantly positive. Column (4) of Table 3 presents the estimation results after incorporating city- and provincial-level control variables. Notably, the impact of short video usage on the mental health of the elderly remains positive and significant. Analysis of the estimation outcomes of the control variables in Column (4) shows that age, gender, individual education level, urban household registration, marital status, family income, development level of the tertiary industry, and number of beds per capita all exhibit a significantly positive impact on the mental health of the elderly. Conversely, government subsidies, family net wealth, and the old-age dependency ratio have a significantly negative impact on the mental health of the elderly. It is imperative to highlight that since Column (4) of Table 3 controls for more confounding factors, this study utilizes Column (4) as the benchmark regression outcome and conducts further analyses, including endogeneity tests and robustness checks, based on this foundation.
Table 3.
Impact of short video usage on the mental health of the elderly people
| Variable | (1) | (2) | (3) | (4) |
|---|---|---|---|---|
| Mental health | Mental health | Mental health | Mental health | |
| Video | 0.023*** | 0.016*** | 0.012** | 0.013*** |
| (0.003) | (0.003) | (0.005) | (0.005) | |
| Age | 0.001*** | 0.001*** | 0.001*** | |
| (0.000) | (0.000) | (0.000) | ||
| Gender | 0.008*** | 0.012*** | 0.011*** | |
| (0.003) | (0.003) | (0.003) | ||
| Education | 0.001*** | 0.001* | 0.001** | |
| (0.000) | (0.000) | (0.000) | ||
| Registration | 0.018*** | 0.009*** | 0.007*** | |
| (0.003) | (0.003) | (0.003) | ||
| Marry | 0.023*** | 0.022*** | 0.021*** | |
| (0.004) | (0.004) | (0.004) | ||
| Smoke | -0.003 | -0.003 | ||
| (0.003) | (0.003) | |||
| Internet | 0.002 | 0.002 | ||
| (0.004) | (0.004) | |||
| Subsidy | -0.009*** | -0.008*** | ||
| (0.003) | (0.003) | |||
| Household size | -0.002** | -0.001 | ||
| (0.001) | (0.001) | |||
| Household asset | 0.000 | -0.001* | ||
| (0.001) | (0.001) | |||
| Household income | 0.008*** | 0.007*** | ||
| (0.001) | (0.001) | |||
| Economy | -0.000 | |||
| (0.004) | ||||
| Population | -0.003 | |||
| (0.004) | ||||
| Tertiary industry | 0.001*** | |||
| (0.000) | ||||
| Housing prices | 0.046 | |||
| (0.031) | ||||
| Beds | 1.019** | |||
| (0.499) | ||||
| Dependency ratio | -0.033** | |||
| (0.013) | ||||
| Province fixed effect | NO | NO | NO | YES |
| Constant | -0.000 | -0.097*** | -0.162*** | -4.367** |
| (0.001) | (0.016) | (0.021) | (2.170) | |
| Observation | 3062 | 3062 | 3062 | 3062 |
| R-squared | 0.015 | 0.066 | 0.084 | 0.104 |
Note: *** p < 0.01, ** p < 0.05, * p < 0.10. Standard errors (clustered at the individual level) are reported in parentheses
Endogeneity
To estimate the impact of short video usage on the mental health of the elderly, this study has diligently endeavored to control for individual-, household-, city-, and provincial-level covariates to the fullest extent possible. However, it is not feasible to control for all potential confounding factors that could concurrently influence the mental health of the elderly and short video usage. In other words, this study may encounter the challenge of omitted variables. Moreover, when surveying older adults, their responses to certain questions may not be entirely precise, indicating potential measurement errors. Furthermore, the use of short videos can influence the mental health of elderly people, and the mental health of elderly people may reciprocally impact short video usage. This may give rise to the issue of mutual causation. To address endogeneity concerns arising from omitted variables, measurement errors, and mutual causation, this study employs two-stage least squares (2SLS) to estimate the impact of short video usage on the mental health of the elderly.
To employ 2SLS to estimate the impact of short video usage on the mental health of the elderly, it is imperative to identify an instrumental variable for the core explanatory variable (short videos). A perfect instrumental variable must satisfy two conditions: first, it must exhibit a correlation with the endogenous explanatory variable, and second, it must meet the criterion of exogeneity. To ensure the validity of the instrumental variable, this study opts for the “Broadband China” pilot policy as the instrumental variable for short video usage. The elderly rely heavily on 4G or 5G networks to access short videos, and the Chinese government has allocated substantial resources to pilot cities under the “Broadband China” initiative, ensuring seamless Internet connectivity for elderly individuals viewing short videos. Consequently, this facilitates the adoption of short videos among the elderly demographic. Hence, a strong correlation exists between the “Broadband China” pilot policy and the usage of short videos among the elderly.
Furthermore, the first-stage estimation results of 2SLS in Column (1) of Table 4 indicates that the estimated coefficient of the “Broadband China” is 0.042, signifying positive significance at the 1% level. Additionally, the F value in the first stage is 58.18, significantly surpassing the critical value of 10, further confirming the robust correlation between the “Broadband China” pilot and short video usage. Moreover, the estimated coefficients obtained by limited information maximum likelihood (LIML) in Column (3) of Table 4 align with those from 2SLS in Column (2). These outcomes suggest that the instrumental variables selected in this study are not weak.
Table 4.
Impact of short video usage on the mental health of the elderly people (considering endogeneity)
| Variable | 2SLS | LIML | |
|---|---|---|---|
| (1) | (2) | (3) | |
| First stage | Second stage | LIML | |
| Video | 0.820*** | 0.820*** | |
| (0.219) | (0.219) | ||
| Instrumental variable | 0.042*** | ||
| (0.011) | |||
| Control variable | YES | YES | YES |
| Province fixed effect | YES | YES | YES |
| Constant | -0.471 | -4.252 | -4.252 |
| (7.792) | (6.430) | (6.430) | |
| Observations | 3062 | 3062 | 3062 |
Note: *** p < 0.01, ** p < 0.05, * p < 0.10. Standard errors (clustered at the individual level) are reported in parentheses
Given the challenge for individuals to influence the “Broadband China” pilot policy, this implies that the instrumental variables utilized in this study are relatively exogenous. Consequently, the analysis demonstrates the appropriateness of selecting the “Broadband China Pilot” as an instrumental variable for short video usage in the elderly. Table 4 reports the estimation results of the impact of short video usage on the mental health of the elderly obtained using the 2SLS model. The estimation results in Column (2) of Table 4 show that after considering endogeneity, the estimated coefficient of the impact of short videos on the mental health of the elderly is 0.82, which is positively significant at the 1% significance level, further demonstrating that the use of short videos can significantly improve the mental health of elderly people.
Robustness check
Robustness check 1: considering the sample self-selection issues
The decision on whether elderly people engage in short video usage is not arbitrary and may be influenced by factors such as education, gender, and geographical location. Therefore, this study encounters a self-selection issue when estimating the impact of short video usage on the mental health of the elderly. Following previous studies [14, 15], this study employs propensity score matching (PSM) to mitigate estimation biases arising from self-selection. In essence, PSM matches each short video user among the elderly with a non-user, ensuring that the only disparity between the two individuals is their short video usage, whereas other characteristics remain largely consistent. The estimation process involves three main steps. Initially, the propensity score for each sample is computed based on individual characteristics, household features, city-level and provincial-level control variables4. Subsequently, individuals in the treatment group (Video = 1) are paired with their counterparts in the control group (Video = 0) with similar propensity scores. Finally, by utilizing successfully matched samples, the disparity in mental health between the elderly short video users and non-users is assessed. This disparity represents the net impact of short video usage on the mental health of the elderly, also known as the average treatment effect on the treated group (ATT).
To use PSM to assess the impact of short video usage on the mental health of the elderly, certain conditions must be satisfied, including balance conditions and the common support assumption. The balance condition stipulates that post-matching, there should be no substantial disparities in covariate means between the treatment (Video = 1) and control groups (Video = 0). Meanwhile, the common support assumption requires sufficient samples for post-matching to compute the average treatment effect. The results of balance tests presented in Table 5 indicate that prior to matching, significant disparities exist in variables such as age, gender, and education between elderly people who use short videos and those who do not. However, post-matching, no significant disparities were observed across all control variables between the elderly in the treatment and control groups. These findings suggest that the balance conditions have been met.
Table 5.
Balance test results
| Variable | (1) | (2) | (3) | (4) | (5) |
|---|---|---|---|---|---|
| Matching status | Mean | T-value | P-value | ||
| Treatment | Control | ||||
| Age | Before | 66.27 | 68.765 | -9.08 | 0.000 |
| After | 66.296 | 66.371 | 0.16 | 0.875 | |
| Gender | Before | 0.597 | 0.515 | 3.46 | 0.001 |
| After | 0.597 | 0.619 | -0.71 | 0.476 | |
| Education | Before | 8.677 | 5.292 | 15.86 | 0.000 |
| After | 8.620 | 8.557 | 0.25 | 0.806 | |
| Registration | Before | 0.632 | 0.461 | 7.31 | 0.000 |
| After | 0.627 | 0.618 | 0.31 | 0.756 | |
| Marry | Before | 0.872 | 0.826 | 2.65 | 0.008 |
| After | 0.871 | 0.873 | -0.12 | 0.904 | |
| Smoke | Before | 0.311 | 0.292 | 0.88 | 0.380 |
| After | 0.309 | 0.331 | -0.78 | 0.434 | |
| Subsidy | Before | 0.362 | 0.422 | -2.55 | 0.011 |
| After | 0.361 | 0.357 | 0.16 | 0.872 | |
| Household size | Before | 3.497 | 3.659 | -1.69 | 0.092 |
| After | 3.500 | 3.451 | 0.43 | 0.670 | |
| Household asset | Before | 1.552 | 0.774 | 7.52 | 0.000 |
| After | 1.285 | 1.339 | -0.37 | 0.712 | |
| Household income | Before | 11.187 | 10.648 | 9.45 | 0.000 |
| After | 11.156 | 11.162 | -0.09 | 0.925 | |
| Economy | Before | 10.969 | 10.865 | 3.45 | 0.001 |
| After | 10.959 | 10.979 | -0.47 | 0.636 | |
| Population | Before | 6.235 | 6.251 | -0.54 | 0.592 |
| After | 6.226 | 6.248 | -0.51 | 0.609 | |
| Tertiary industry | Before | 53.938 | 52.451 | 3.02 | 0.003 |
| After | 53.751 | 53.618 | 0.19 | 0.847 | |
| Housing prices | Before | 9.164 | 9.082 | 3.58 | 0.000 |
| After | 9.154 | 9.159 | -0.16 | 0.869 | |
| Beds | Before | 4.158 | 4.155 | 0.49 | 0.623 |
| After | 4.158 | 4.160 | -0.30 | 0.764 | |
| Dependency ratio | Before | 18.444 | 18.349 | 0.60 | 0.546 |
| After | 18.404 | 18.487 | -0.40 | 0.688 | |
In addition to meeting the balance condition, the application of PSM requires satisfying the common support assumption. To assess the fulfillment of this assumption, this study presents density plots of the propensity scores for the treatment and control groups before and after matching. Examination of the density plots depicted in Figs. 1 (before matching) and 2 (after matching) reveals minimal disparity in the distribution range of propensity score values between the treatment and control groups post-matching. Consequently, we can deduce that the common support assumption is upheld. These test outcomes affirm the suitability of employing PSM to investigate the influence of short videos on the mental health of the elderly.
Fig. 1.
Propensity score kernel density diagram of treatment and control groups before matching
Fig. 2.
Propensity score kernel density diagram of treatment and control groups after matching
Referring to previous studies [6], this study employs various matching techniques, including radius matching, kernel matching, K-nearest neighbor matching (K = 1 and K = 4), and local linear regression matching, to compute the average treatment effect. Table 6 presents the average treatment effects derived from different matching methods. The estimation outcomes reveal that the average treatment effect value obtained using each matching method is positively significant. This suggests that the use of short videos can substantially enhance the mental health of the elderly, which aligns with earlier estimation findings.
Table 6.
Impact of short video usage on the mental health of the elderly people (considering self-selection)
| Matching method | (1) | (2) | (3) | (4) | (5) |
|---|---|---|---|---|---|
| Radius matching | Kernel matching |
Nearest neighbor matching (k = 1) | Nearest neighbor matching (k = 4) | Local linear matching | |
| ATT | 0.0121*** | 0.0143*** | 0.0145*** | 0.0129*** | 0.0144*** |
| (0.0040) | (0.0035) | (0.0047) | (0.0038) | (0.0047) | |
| Observations | 3062 | 3062 | 3062 | 3062 | 3062 |
Note: *** p < 0.01, ** p < 0.05, * p < 0.10. Standard errors (clustered at the individual level) are reported in parentheses
Robustness check 2: replacing the measurement method of elderly mental health
In the aforementioned analysis, we utilize factor analysis to gauge the mental health of the elderly. Referring to previous studies [6], the mental health of the elderly is assessed using both direct summation and principal component analysis. Prior to employing the principal component analysis method for mental health assessment, this study conducts Kaiser-Meyer-Olkin (KMO) and Bartlett’s sphericity tests. The results of these tests indicate that the KMO values for each sub-variable in the CES-D scale exceed 0.8, and Bartlett’s sphericity test rejects the null hypothesis of no correlation between the sub-indicators in the scale at the 1% significance level. These outcomes affirm the appropriateness of utilizing principal component analysis to assess the mental health of the elderly. The estimation results in Columns (1) and (2) of Table 7 demonstrate that regardless of whether mental health, estimated through direct summation or principal component analysis, serves as the dependent variable, the estimated coefficients of the impact of short video usage on the mental health of elderly people remain positively significant. These further underscores the fact that the use of short videos among the elderly significantly enhances their mental health.
Table 7.
Robustness test results
| Variable | (1) | (2) | (3) | (4) |
|---|---|---|---|---|
| Mental health | Mental health | Mental health | Mental health | |
| Video | 12.966*** | 11.610*** | 4.152*** | 0.398** |
| (3.459) | (3.095) | (0.134) | (0.167) | |
| Control variable | YES | YES | YES | YES |
| Province fixed effect | YES | YES | YES | YES |
| Constant | -40.996 | -37.018 | -10.407 | -54.710 |
| (102.520) | (91.476) | (32.856) | (66.245) | |
| Observations | 3062 | 3062 | 3062 | 3062 |
Note: *** p < 0.01, ** p < 0.05, * p < 0.10. Standard errors (clustered at the individual level) are reported in parentheses
Robustness check 3: substituting the model
This study transforms the dependent variable into dummy variables and employs alternative models to investigate the influence of short videos on the mental health of the elderly. If an elderly individual’s mental health level exceeds the mean of all elderly participants, the dummy variable is assigned a value of 1; otherwise, it is assigned a value of 0. Given the dummy nature of the dependent variable, this study further employs the IV-probit and logit models to investigate the impact of short video usage on the mental health of elderly people. The estimation outcomes in Columns (3)–(4) of Table 7 reveal that the effect of short video usage on the mental health of the elderly remains positive and significant. These further underscores the fact that the use of short videos can enhance the mental health of the elderly.
Heterogeneity effects
Heterogeneity in educational attainment
Recognizing that the influence of short videos on the mental health of various elderly people may not be uniform, this study stratifies the sample into groups with differing levels of education to examine the effects of short videos on the mental health of elderly people with distinct educational backgrounds. To categorize the sample, an individual is classified as relatively well educated if their educational attainment surpasses the mean level of education among all elderly participants; otherwise, they are considered to have a low educational level. Furthermore, this study employs the 2SLS to explore the impact of short video usage on the mental health of older adults with varying educational backgrounds. The estimation results in Columns (1) and (2) of Table 8 reveal that the estimated coefficients of the effect of short video usage on the mental health of highly and less educated elderly people are 0.735 (P < 0.01) and 1.202 (P < 0.1), respectively. This signifies that short video use has a substantial positive influence on the mental health of elderly people across different educational backgrounds. Comparing the estimated coefficients in Columns (1) and (2), it is evident that short videos have a more pronounced impact on the mental health of older individuals with lower levels of education.
Table 8.
Heterogeneity analysis results
| Variable | (1) | (2) | (3) | (4) |
|---|---|---|---|---|
| Education | Gender | |||
| High | Low | Female | Male | |
| Video | 0.735*** | 1.202* | 1.020* | 0.790*** |
| (0.219) | (0.686) | (0.594) | (0.247) | |
| Control variable | YES | YES | YES | YES |
| Province fixed effect | YES | YES | YES | YES |
| Constant | -4.995 | -13.160 | -4.519 | -1.627 |
| (7.516) | (9.933) | (6.864) | (9.758) | |
| Observations | 1941 | 1121 | 1440 | 1622 |
Note: *** p < 0.01, ** p < 0.05, * p < 0.10. Standard errors (clustered at the individual level) are reported in parentheses
Gender heterogeneity
To explore the diverse effects of short video usage on the mental health of older adults across different genders, this study employs 2SLS to investigate its impact on both female and male elderly people. The results presented in Columns (3) and (4) of Table 8 indicate that the estimated coefficients of the influence of short video usage on the mental health of elderly female and male populations are both positive and statistically significant. This suggests that short video usage affects the mental health of both female and male elderly groups equally. Upon comparing the estimated coefficients of short video usage in Columns (3) and (4), it becomes apparent that the coefficients in Column (3) are notably larger than those in Column (4). These findings suggest that the impact of short video usage on the mental health of elderly women surpasses that of elderly men, indicating a more pronounced effect in the former.
Mechanism analysis
Leisure consumption
This study employs the mediating effect model to elucidate the mechanism through which short videos influence the mental health of the elderly by affecting their leisure consumption. Leisure consumption is gauged as an intermediary variable represented by the aggregate sum of cultural and entertainment expenditures alongside tourism expenses among the elderly. A higher value indicates greater leisure consumption. Given the relatively large magnitude of leisure consumption values, this study applies a logarithmic transformation to the analysis. The estimation results in Column (1) of Table 9 reveal a positive and statistically significant estimated coefficient for the impact of short video usage on leisure consumption, indicating a significant enhancement in leisure consumption due to short video usage. This finding aligns with the conclusions of previous studies [16]. Furthermore, the estimation results in Column (2) of Table 9 show that the estimated coefficient of the impact of leisure consumption on the mental health of elderly people is 0.0003, which is positively significant, indicating that an increase in leisure consumption can significantly enhance the mental health of elderly people. These outcomes collectively affirm that short video usage can improve the mental health of the elderly by fostering their engagement in leisure activities.
Table 9.
The mediating effect of leisure consumption test results
| Variable | (1) | (2) |
|---|---|---|
| Leisure consumption | Mental health | |
| Video | 0.696*** | 0.015*** |
| (0.164) | (0.004) | |
| Leisure consumption | 0.0003** | |
| (0.00013) | ||
| Control Variable | YES | YES |
| Province fixed effect | YES | YES |
| Constant | 144.788 | -4.726** |
| (100.986) | (2.189) | |
| Observations | 3062 | 3062 |
Note: *** p < 0.01, ** p < 0.05, * p < 0.10. Standard errors (clustered at the individual level) are reported in parentheses
Intergenerational relationships
To investigate whether short videos can enhance the intergenerational relationship between elderly individuals and their children, thereby influencing improvements in their mental health, this study employs a mediation effect model for validation. The CFPS included inquiries about intergenerational relationships between parents and children. Specifically, respondents were asked, “How would you describe your relationship with your children?” with options ranging from (a) not very close, (b) somewhat close, (c) average, (d) close to, (e) very close. Each response was assigned a value ranging from 1 to 5, with higher scores indicating stronger bonds. Considering that elderly individuals may have multiple children, this study calculates the average relationship score between the elderly and their children, where a higher value signifies a better intergenerational relationship.
According to the estimation results presented in Column (1) of Table 10, the coefficient estimate for the effect of short videos on the intergenerational relationship between the elderly and their children is 0.13, signifying statistical significance at the 1% level. This indicates that the use of short videos significantly enhances intergenerational relationships between the elderly and their children. Furthermore, the estimation results in Column (2) of Table 10 show a positively significant coefficient for the influence of intergenerational relationships on mental health of elderly individuals. This finding implies that strong intergenerational bonds can positively affect elderly people’s mental well-being. In conclusion, our findings support the notion that short videos have the potential to strengthen the bond between the elderly and their children, thereby enhancing their mental health.
Table 10.
The mediating effect of intergenerational relationships test results
| Variable | (1) | (2) |
|---|---|---|
| Intergenerational relationships | Mental health | |
| Video | 0.130*** | 0.012*** |
| (0.033) | (0.004) | |
| Intergenerational relationships | 0.018*** | |
| (0.002) | ||
| Control Variable | YES | YES |
| Province fixed effect | YES | YES |
| Constant | 0.054 | -4.758** |
| (22.084) | (2.178) | |
| Observations | 2883 | 2883 |
Note: *** p < 0.01, ** p < 0.05, * p < 0.10. Standard errors (clustered at the individual level) are reported in parentheses
Discussion
The advent of the new technological revolution, epitomized by Internet technology, has brought substantial changes in socio-economic dynamics and individuals’ lifestyles, profoundly influencing the mental health of the elderly. At the macro level, the widespread adoption of the Internet has spurred the emergence of new economies, such as e-commerce, fostering industrial productivity enhancement, augmenting household incomes, and narrowing the spatial gap between the elderly and their offspring [17]. At the micro level, the Internet enables older adults to recognize the detrimental effects of unhealthy behaviors and cultivate healthier lifestyles [18]. Given its impact on household incomes, personal behaviors, and emotional states, the Internet may also influence health. A plethora of scholarly inquiries have delved into the nexus between Internet use and health, with the majority indicating a positive association between Internet use and the health of older adults [19–23].
Owing to the unique characteristics of the elderly demographic, they may be susceptible to Internet addiction, which could potentially impact their health. First, akin to adolescents, the elderly may exhibit weaker self-discipline, rendering them vulnerable to becoming “Internet-addicted elderly individuals,” thereby affecting their mental health. Second, Internet usage could disrupt the elderly’s daily routines and encroach on their sleep patterns. Finally, the abundance of information available online increases the elderly’s susceptibility to deception and potential financial losses. Indeed, some studies have found that Internet use can detrimentally affect the mental health of older adults [24]. Given that short videos are a by-product of Internet development, insights into their impact and mechanisms on the mental health of elderly people can be gleaned from pertinent literature examining “The Impact of Internet Development on the Health of the Elderly.” As is evident from the aforementioned discussion, there is no consensus regarding the impact of the Internet on older adults’ mental health. Consequently, the potential impact of short video usage on the mental health of the elderly remains uncertain.
However, there is a lack of consensus on the influence of short videos on individual health. Numerous studies have explored the relationship between short video usage and health outcomes [25–30]. Most of these investigations have demonstrated the detrimental effects of short video use on adolescent mental health. For instance, previous studies found that excessive use of short videos increases the risk of depression among adolescents [27, 28]. Furthermore, the irrational use of short video social media may elevate the risk of self-harm and suicidal behavior among teenagers [30]. Contrary to these negative findings, some studies have reported positive effects of short video use on adolescent mental health [31]. Furthermore, some studies have not distinguished between age groups when exploring the effects of social media on health outcomes [32, 33]. Diverging from these prior inquiries, this study specifically focuses on evaluating the influence of short videos on the mental health of the elderly.
As the global population ages, considerable research is being devoted to investigating the influence of social media use on the mental health of older adults. Numerous studies have documented the significant positive impact of social media use on the mental health of the elderly [34–37]. For instance, previous studies found that during the COVID-19 pandemic, the use of short videos helped reduce loneliness and improve the mental health of the elderly [37]. However, the aforementioned studies predominantly focus on developed nations, such as the United States. In contrast, this study adopts a distinct perspective by examining China, a rapidly developing country, thus contributing to a more comprehensive understanding of the relationship between social media usage and the mental health of older adults.
Studies have also investigated the influence of social media on the mental health of elderly people in China. For instance, previous studies explored the effects of WeChat, a popular social media platform, on both the physical and mental health of older adults, using data from the 2017 Chinese General Social Survey [38]. Their findings indicated significant improvements in both physical and mental health among the elderly who use WeChat. In contrast to the research mentioned above, this study focuses specifically on the impact of short video social media use on the mental health of older adults. Reasonable use of short videos can improve the mental health of the elderly; however, addiction to short video social media may have adverse effects. Other studies have found that short video addiction not only heightens an individual’s negative emotions but also affects their spouse’s emotional well-being [39]. Previous studies investigated the relationship between short video usage and depression levels in elderly people [40]. The study found that the excessive use of short video social media may negatively impact the mental health of the elderly. However, some studies do not address the potential endogeneity in the effects of short video use on the mental health of older adults [39, 40]. In contrast to these studies [39, 40], our study employs a large-scale dataset and addresses endogeneity concerns in the empirical analysis. We employ instrumental variables derived from the “Broadband China” pilot initiative and utilize a 2SLS regression to examine the impact of short video usage on the mental health of older adults. This study offers a more accurate reflection of the causal relationship between social media usage and health. The baseline regression results demonstrate that the use of short videos significantly enhances the mental health of the elderly. These findings align with the conclusions drawn from the majority of existing literature exploring the impact of social media on the health of older adults.
The findings of the heterogeneity analysis reveal that short videos exhibit a more pronounced positive effect on the mental health of elderly people with lower levels of education. This phenomenon may stem from the increased susceptibility to loneliness and social isolation among older adults with limited education [41]. Short video social media platforms offer user-friendly interfaces with low entry barriers, enabling easier engagement among elderly individuals with lower educational attainment. Consequently, these individuals may find short videos to be a convenient means of social interaction, thereby enhancing their frequency of social engagement. Conversely, elderly individuals with higher levels of education may exhibit a preference for traditional media formats such as books, newspapers, or in-depth journalistic programs. The relatively brief and entertaining nature of short videos may not align with the media preferences of this demographic. Furthermore, individuals with higher educational backgrounds often prioritize the depth and thoroughness of information. As short videos typically offer concise content with limited depth, they may struggle to resonate with this segment of the elderly population.
The results of the heterogeneity analysis also indicate that short videos have a lower impact on the mental health of elderly males compared to females. This could be attributed to the general observation that women tend to be more adept at expressing and discerning emotions than men. Short videos often leverage emotional stimuli, such as narrative arcs, musical accompaniments, and visual effects, to evoke emotional responses from viewers. Given their heightened emotional sensitivity, women may be more predisposed to profound emotional reactions to such content, rendering them more susceptible to the influence of short videos. Moreover, women typically demonstrate a greater inclination toward visual aesthetics and sensory experiences. Short videos, as a visually immersive medium, capitalize on captivating imagery, vibrant colors, and resonant music to captivate viewers. These elements may resonate more deeply with female audience members, capturing their attention and eliciting emotional responses more effectively.
The results of the mechanism analysis indicate that social media can influence the mental health of the elderly by augmenting their leisure consumption. Indeed, older adults’ use of short videos can affect leisure consumption across various dimensions. First, certain short video platforms offer premium entertainment content that requires users to pay for access. Elderly individuals who watch short videos are more likely to purchase related offerings, thus enhancing their entertainment experiences. Second, some short video applications integrate offline events and experiences, such as brand collaborations featuring live events, performances, and exhibitions. Elderly individuals may be willing to pay to enrich their cultural and entertainment experiences, resulting in an increase in leisure consumption. Furthermore, short video content often incorporates product or service endorsements, stimulating interest among elderly viewers, who may subsequently make related purchases, thereby augmenting their leisure consumption. Finally, short videos frequently showcase cultural and tourism-related content, inspiring older viewers to increase their leisure consumption of cultural experiences and travel. Consequently, the use of short videos by the elderly may contribute to an increase in leisure consumption levels.
The reasons why an increase in leisure consumption can improve the mental health of the elderly are as follows: First, heightened leisure consumption affords older adults’ greater opportunities to partake in various leisure pursuits such as movie-watching, cultural engagements, club participation, and more. These leisure activities serve as effective antidotes for loneliness among the elderly, thereby bolstering their overall life satisfaction. Furthermore, prior research has demonstrated that physical activities such as walking, swimming, and fitness regimens play pivotal roles in preserving the physical and mental health of older individuals [42, 43]. Moreover, engaging in leisure activities enables elderly people to unearth and cultivate new interests and hobbies, injecting vibrancy into their lives. Consequently, augmentation of leisure consumption is conducive to enhancing the mental health of the elderly. Indeed, a wealth of literature has corroborated the positive impact of increased leisure activities on residents’ mental health [44–46].
The results of the mechanism analysis further demonstrate that the use of short videos can impact intergenerational relationships between the elderly and their children, consequently influencing their mental well-being. This can be attributed to the natural progression of aging in which interpersonal circles tend to shrink, leading to the simplification of complex relationships, with intergenerational bonds becoming increasingly vital in later life. Strong intergenerational ties offer crucial social support for the elderly, mitigating feelings of loneliness, anxiety, and depression, thereby enhancing their mental health. Additionally, fostering positive intergenerational relationships encourages elderly individuals to engage in social activities, thereby promoting active and fulfilling lifestyles. Numerous studies have corroborated the positive impact of robust intergenerational relationships on mental health in older adults [47–49].
The use of short videos can enhance intergenerational relationships between elderly individuals and their children because short videos serve as a simple and direct means of communication, facilitating easier interaction across generations. By sharing engaging video content, individuals can overcome age barriers, fostering communication and interaction. Additionally, elderly individuals can establish connections by engaging in activities such as sharing, commenting, and discussing short videos with their children, thus enhancing communication and fostering mutual understanding. Consequently, the use of short videos fosters harmonious intergenerational relationships between the elderly and their children.
This study has certain limitations that warrant acknowledgment. First, while this study diligently controls for a range of covariates such as age, gender, and education level, it remains constrained by the inability to account for all potential confounding factors that could influence both short video usage and the mental health of elderly people. Second, this investigation only delves into the short-term ramifications of short video usage on the mental health of elderly people, failing to examine its long-term effects. Future research endeavors, with access to comprehensive datasets, could explore the long-term impact of short video engagement on the mental health of older adults. Finally, although the instrumental variables selected in this study are relatively exogenous, they are not perfect. Therefore, using instrumental variables and the 2SLS model to estimate the impact of short video use on the mental health of the elderly may result in minor biases. Future research could explore the impact of these exogenous shocks on the mental health of the elderly by leveraging quasi-natural experiments that may have exogenous effects on short videos.
The findings of this study have policy implications. It is evident that the use of short videos by the elderly positively impacts their mental health. Consequently, governmental intervention is warranted to promote digital literacy among the elderly population, offering pertinent education, training, and resources to facilitate their safe and judicious use of short video social media platforms. Concurrently, governmental support for the integration of short video platforms into healthcare services is imperative. This entails leveraging such platforms to disseminate medical knowledge and deliver health-related guidance, thereby providing elderly people with comprehensive healthcare information and services. Additionally, the government could focus on promoting the development of short video-based mutual aid communities for the elderly. Encouraging elderly individuals to participate in such communities facilitates the exchange of experiences and mutual support, thereby promoting social interactions and alleviating feelings of loneliness within this demographic. Finally, the government should prioritize enhancing network security measures, augmenting elderly individuals’ awareness of online safety, and fortifying privacy protection on short video platforms. Such measures are essential to safeguard the elderly from online fraud and ensure their secure utilization of digital platforms.
Conclusion
Enhancing the mental health of the elderly has significant implications for their welfare and stable development of the economy and society. Drawing on data from the 2020 CFPS, this study investigates the influence of short video usage on the mental health of the elderly. These findings indicate a noteworthy improvement in the mental health of elderly people attributable to their engagement with short videos. To address potential endogeneity concerns, this study employs the “Broadband China” pilot policy as an instrumental variable for short video usage and applies 2SLS to evaluate the impact of short video usage on mental health. The estimation results, accounting for endogeneity, indicate that short video usage among the elderly can enhance their mental health. To ensure the robustness of these findings, we conduct sensitivity analyses by exploring alternative models and dependent variables, yielding consistent conclusions. Heterogeneity analyses reveal that the positive impact of short video usage on mental health is particularly pronounced among women and elderly individuals with lower educational attainment. Furthermore, the mechanism analysis results indicate that the use of short videos can enhance the mental health of the elderly by influencing intergenerational relationships and leisure consumption.
Acknowledgements
I would like to thank Jiahui Xia, Ruidong Sun for their helpful remarks.
Abbreviations
- CFPS
China Family Panel Studies
- 2SLS
Two-Stage Least Squares
- CES-D
Center for Epidemiologic Studies Depression
- LIML
Limited Information Maximum Likelihood
- PSM
Propensity Score Matching
- ATT
Average Treatment Effect on the Treated
- KMO
Kaiser-Meyer-Olkin
Author contributions
R.Z collects the data, designs the model, and finishes the empirical analysis. Y.S collects the data and writes the manuscript. Z.L writes the manuscript. X.H writes the manuscript. All authors read and approved the final manuscript.
Funding
This research was funded by the Natural Science Foundation of Sichuan (Grant/Award Number: 2024NSFSC1091), Guangdong Basic and Applied Basic Research Foundation (No. 2022A1515111004), and Guangdong Planning Project of Philosophy and Social Science (No. GD22YYJ05).
Data availability
The datasets generated and analyzed during the current study were derived from the China Family Panel Studies (CFPS). They are opened to everyone. Researchers who want to use these data can visit http://www.isss.pku.edu.cn/cfps/.
Declarations
Ethics approval and consent to participate
Not applicable.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Data Sources: https://www.stats.gov.cn/sj/pcsj/rkpc/d7c/.
Following the World Health Organization’s definition, we classify residents aged 60 and above as elderly.
Data Sources: https://www.cbbpa.org.cn/Detail/889_5150.
Given that individuals who use short videos are inherently Internet users, the variable for Internet usage is excluded when applying PSM to estimate the effect of short video use on mental health of the elderly.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Yang S, Huang L, Zhang Y, et al. Unraveling the links between active and passive social media usage and seniors’ loneliness: a field study in aging care communities. Internet Res. 2021;31(6):2167–89. [Google Scholar]
- 2.Rasmussen EE, Punyanunt-Carter N, LaFreniere JR, et al. The serially mediated relationship between emerging adults’ social media use and mental well-being. Comput Hum Behav. 2020;102:206–13. [Google Scholar]
- 3.Liu C, Ma J. Social media addiction and burnout: the mediating roles of envy and social media use anxiety. Curr Psychol. 2020;39(6):1883–91. [Google Scholar]
- 4.Zhong B, Jiang Z, Xie W, et al. Association of social media use with mental health conditions of nonpatients during the COVID-19 outbreak: insights from a national survey study. J Med Internet Res. 2020;22(12):e23696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Liao L, Du M, Chen Z. Air pollution, health care use and medical costs: evidence from China. Energy Econ. 2021;95:105132. [Google Scholar]
- 6.Zhang R, Feng D, Xia J, et al. The impact of household wealth gap on individual’s mental health. BMC Public Health. 2023;23(1):1936. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1(3):385–401. [Google Scholar]
- 8.Gong J, Lu Y, Xie H. The average and distributional effects of teenage adversity on long-term health. J Health Econ. 2020;71:102288. [DOI] [PubMed] [Google Scholar]
- 9.Li H, Li J, Lu Y, et al. Do housing regulations affect child development? Evidence and mechanisms. J Public Econ. 2023;227:104995. [Google Scholar]
- 10.Liao L, Du M. How digital finance shapes residents’ health: evidence from China. China Econ Rev. 2024;87:102246. [Google Scholar]
- 11.Liu P, Teng M, Han C. How does environmental knowledge translate into pro-environmental behaviors? The mediating role of environmental attitudes and behavioral intentions. Sci Total Environ. 2020;728:138126. [DOI] [PubMed] [Google Scholar]
- 12.Liu P, Han C, Liu X, et al. From smoke to smiles: quantifying the happiness benefits of household cooking energy transition. J Environ Manage. 2024;368:122203. [DOI] [PubMed] [Google Scholar]
- 13.Wu S, Zhang R, Wang C, et al. The impact of natural disasters on rural household wealth: Micro evidence from China. Front Environ Sci. 2022;10:993722. [Google Scholar]
- 14.Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70(1):41–55. [Google Scholar]
- 15.Zhang R, Zhang Y, Xia J. Impact of mobile payment on physical health: evidence from the 2017 China household finance survey. Front Public Health. 2022;10:963234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Phucharoen C, Wichupankul S, Sangkaew N, et al. Social media sites and nightlife tourism: an empirical study of tourist decision-making on nightlife entertainment consumption. Int J Tourism Cities. 2023;9(3):751–70. [Google Scholar]
- 17.Bartel A, Ichniowski C, Shaw K. How does information technology affect productivity? Plant-level comparisons of product innovation, process improvement, and worker skills. Q J Econ. 2007;122(4):1721–58. [Google Scholar]
- 18.Campbell RJ. Nolfi. Teaching elderly adults to use the internet to access health care information: before-after study. J Med Internet Res. 2005;7(2):e128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Karavidas M, Lim NK. Katsikas. The effects of computers on older adult users. Comput Hum Behav. 2005;215:697–711. [Google Scholar]
- 20.Cohall AT, Nye A, Moon-Howard J, et al. Computer use, internet access, and online health searching among Harlem adults. Am J Health Promotion. 2011;25(5):325–33. [DOI] [PubMed] [Google Scholar]
- 21.Van Boekel LC, Peek STM, Luijkx KG. Diversity in older adults’ use of the internet: identifying subgroups through latent class analysis. J Med Internet Res. 2017;19(5):e180. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Lam SS, Man S, Jivraj, Scholes S. Exploring the relationship between internet use and mental health among older adults in England: longitudinal observational study. J Med Internet Res. 2020;22:7:e15683. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Yuan H. Internet use and mental health problems among older people in Shanghai, China: the moderating roles of chronic diseases and household income. Aging Ment Health. 2021;254:657–63. [DOI] [PubMed] [Google Scholar]
- 24.Nie NH, Erbring L. Internet and society: a preliminary report. IT Soc. 2002;1(1):275–83. [Google Scholar]
- 25.Ye JH, Wu YT, Wu YF, et al. Effects of short video addiction on the motivation and well-being of Chinese vocational college students. Front Public Health. 2022;10:847672. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Linlin W, Wanyu H, Yuting L, et al. Research on the mechanism of short video information interaction behavior of college students with psychological disorders based on grounded theory. BMC Public Health. 2023;23(1):2256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Zhang D, Yang Y, Guan M. A cross-lagged analysis of the relationship between short video overuse behavior and depression among college students. Front Psychol. 2024;15:1345076. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Qu D, Liu B, Jia L, et al. The longitudinal relationships between short video addiction and depressive symptoms: a cross-lagged panel network analysis. Comput Hum Behav. 2024;152:108059. [Google Scholar]
- 29.Liu M, Zhuang A, Norvilitis JM, et al. Usage patterns of short videos and social media among adolescents and psychological health: a latent profile analysis. Comput Hum Behav. 2024;151:108007. [Google Scholar]
- 30.Yu Z, Zhu X, Li Y. The association between problematic short video use and suicidal ideation and self-injurious behaviors: the mediating roles of sleep disturbance and depression. BMC Public Health. 2024;24(1):1689. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Yang Y, Wang H, Sha W et al. Short Video-Based Mental Health Intervention for Depressive Symptoms in Junior High School students: a Cluster Randomized Controlled Trial. Psychol Res Behav Manage. 2023;4169–81. [DOI] [PMC free article] [PubMed]
- 32.Cheng X, Su X, Yang B, et al. Understanding users’ negative emotions and continuous usage intention in short video platforms. Electron Commer Res Appl. 2023;58:101244. [Google Scholar]
- 33.Zuo X, Wang R, Hong Z. Does using short video apps impacts life satisfaction: a perspective from psycho-social mechanism. Curr Psychol. 2024;1–15.
- 34.Blomfield Neira CJ, Barber BL. Social networking site use: linked to adolescents’ social self-concept, self‐esteem, and depressed mood. Australian J Psychol. 2014;66(1):56–64. [Google Scholar]
- 35.Zheng Z, Sun N, Chen Y, et al. Social media use and mental health of urban residents during China’s second COVID-19 outbreak. Front Public Health. 2022;10:1016849. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Chopik WJ, Cyberpsychology. Behav Social Netw. 2016;19(9):551–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Pandele VF, Tucmeanu A, Antohi ME, et al. The effect of using Facebook on elderly people during COVID-19 pandemic. Archiv Euromedica. 2021;11(5):69–71. [Google Scholar]
- 38.Fu L, Xie Y. The effects of social media use on the health of older adults: an empirical analysis based on 2017 Chinese general social survey//Healthcare. MDPI. 2021;9(9):1143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Deng J, Wang M, Mu W, et al. The relationship between addictive use of short-video platforms and marital satisfaction in older Chinese couples: an asymmetrical dyadic process. Behav Sci. 2024;14(5):364. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Wen X, Zhou Y, Li Y et al. Perceived overload on short video platforms and its influence on Mental Health among the Elderly: a Moderated Mediation Model. Psychol Res Behav Manage. 2024;2347–62. [DOI] [PMC free article] [PubMed]
- 41.Bjelland I, Krokstad S, Mykletun A, et al. Does a higher educational level protect against anxiety and depression? The HUNT study. Soc Sci Med. 2008;66(6):1334–45. [DOI] [PubMed] [Google Scholar]
- 42.Deslandes A, Moraes H, Ferreira C, et al. Exercise and mental health: many reasons to move. Neuropsychobiology. 2009;59(4):191–8. [DOI] [PubMed] [Google Scholar]
- 43.Grasdalsmoen M, Eriksen HR, Lønning KJ, et al. Physical exercise, mental health problems, and suicide attempts in university students. BMC Psychiatry. 2020;20(1):1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Kekäläinen T, Wilska TA, Kokko K. Leisure consumption and well-being among older adults: does age or life situation matter? Appl Res Qual Life. 2017;12(3):671–91. [Google Scholar]
- 45.Nimrod G, Kleiber DA, Berdychevsky L. Leisure in coping with depression. J Leisure Res. 2012;44(4):419–49. [Google Scholar]
- 46.Nagata S, McCormick B, Piatt J. Leisure behavior and sense of coherence in the context of depression. J Community Psychol. 2020;48(2):283–301. [DOI] [PubMed] [Google Scholar]
- 47.Chen X, Silverstein M. Intergenerational social support and the psychological well-being of older parents in China. Res Aging. 2000;22(1):43–65. [Google Scholar]
- 48.Chunkai L, Shan J, Xinwen Z. Intergenerational relationship, family social support, and depression among Chinese elderly: a structural equation modeling analysis. J Affect Disord. 2019;248(1):73–80. [DOI] [PubMed] [Google Scholar]
- 49.Zheng R, Yu M, Huang L, et al. Effect of intergenerational exchange patterns and intergenerational relationship quality on depressive symptoms in the elderly: an empirical study on CHARLS data. Front Public Health. 2022;10:1009781. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets generated and analyzed during the current study were derived from the China Family Panel Studies (CFPS). They are opened to everyone. Researchers who want to use these data can visit http://www.isss.pku.edu.cn/cfps/.