Increasing self-esteem among left-behind children in China: a cross-temporal meta-analysis from 2006 to 2020

Abstract

Background

Self-esteem serves as a fundamental indicator for assessing the mental health of left-behind children in China. Previous studies have placed more emphasis on investigating its current situation, while overlooking its dynamic changes with social development.

Methods

This research project used the approach of cross-temporal meta-analysis to analyze 72 papers that used the Rosenberg Self-esteem Scale, ranging from 2006 to 2020(date of data collection) and covering 29,835 left-behind children, to delve into the diachronic changes and influencing factors of self-esteem of left-behind children over the years.

Results

The outcomes demonstrated that: (1) The self-esteem scores of Chinese left-behind children increased from 30.74 in 2006 to 32.7 in 2020. (2) The level of self-esteem among left-behind children was significantly correlated with social indicators (economic conditions and educational level) for the actual year, 3 years prior, or 5 years prior to the self-esteem data collection.

Conclusions

The level of self-esteem of left-behind children has been rising between 2006 and 2020. In addition, it is noteworthy that social development had a significant influence on predicting changes in the self-esteem of left-behind children.

Introduction

Since China’s reform and opening up, the process of urbanization has been accelerating, and a considerable rural labor population have poured into cities to seek better jobs and higher incomes. However, due to the lack of stable living conditions, many of these migrants are unable to bring their children along, resulting in a group commonly referred to as ‘left-behind children’. Left-behind children are minors under the age of 18 who remain at their registered place of residence while one or both of their parents are away for work [1]. Based on the 2018 statistics provided by the Ministry of Civil Affairs, the number of left-behind children in China still remains close to 7 million, making them a significant and undeniable population that demands attention. Growing up without the presence of their parents, these children face various challenges in their development, including a lack of parental care, education, guidance, and overall support [2]. Numerous studies have demonstrated that such unfavorable family environments significantly increase the likelihood of mental health problems among left-behind children, compared to those who live with their parents [3, 4]. According to the White Paper on the Psychological Status of China’s left-behind Children released in 2015, low self-esteem has become a serious problem affecting the mental health of these children. Left-behind children with low self-esteem are more vulnerable to anxiety and depression [5], and have more problem behaviors [6]. Therefore, the self-esteem has gradually become a mental health indicator that the government and scholars focus on.

Self-esteem signifies the individual’s perception of their own ability and value as well as the degree of satisfaction with the overall self [7]. The focus on left-behind children is relatively late in China, and it was not until 2007 that there was an empirical investigation into the self-esteem level. The self-esteem scale compiled by Rosenberg is mainly used as measuring tool in China. Although abundant research results have been obtained, these studies have not formed a unified opinion. Some studies have indicated that left-behind children tend to have lower self-esteem compared to their non-left-behind counterparts [8, 9]. However, some studies have found that there is no significant disparity in self-esteem between left-behind children and their non-left-behind counterparts [10]. Surprisingly, certain studies have even shown that left-behind children exhibit higher self-esteem than non-left-behind children [11]. The reason for the inconsistency of research results may be due to the difference of measurement indicators, samples and statistical methods adopted in each study [12]. In addition, self-esteem, as a product of the social relationship network, may also be affected by the social environment [13]. Especially in the period of social transition, our society has experienced significant changes across various domains, including the rapid economic growth and the widespread promotion of education. Moreover, in recent years, the Party and the government have recognized and prioritized the psychological issues faced by left-behind children. In order to promote the healthy growth of them, the state has issued relevant policies. For example, in 2016, The State Council issued “Opinions on Strengthening the Care and Protection of left-behind Children in Rural areas”. Since then, a total of 26 provinces, including Shandong, Fujian and Zhejiang, have put forward specific measures in combination with local actual conditions, with the goal of enhancing the support and protection available to left-behind children. Therefore, in such a social environment, it is worth examining whether the self-esteem levels of left-behind children undergo any changes over the years. However, there are few studies to depict the longitudinal trend of left-behind children’s self-esteem. Therefore, this study attempts to make clear the changing trend of left-behind children through the utilization of cross-temporal meta-analysis, with a view to reaching more general conclusion, so as to help the construction of the mental health service system for left-behind children.

Cross-temporal meta-analysis is used to explore continuous changes of a certain psychological variable over the years to make up for the “age effect” problem that cannot be solved by general methods [14]. Through this method, we can summarize all empirical studies related to a certain psychological variable in a period of time, and extract the effective research results according to the year. Then, by analyzing the relationship between the research results and the year, we can describe the track of psychological variables changing with the year. Chinese scholars have used this approach to explore the evolving dynamics of mental health [15, 16], life satisfaction [17], well-being [18] and other psychological variables of different groups in China, and obtained rich and meaningful research results.

Cross-temporal meta-analysis can also shed light on the relationship between the changes of psychological variables and social macro factors by using the lag correlation analysis [19, 20]. Based on relevant studies, the social factors that may affect self-esteem of left-behind children mainly focus on China’s economic conditions and educational level. First, economic conditions play a fundamental role in psychological changes. Studies have found that income and quality of life can positively predict self-esteem [21, 22]. The generation of children left behind is ultimately due to the unbalanced development of China’s economy after the reform and opening up, resulting in a surge in the demand for urban labor force, which leads to the influx of rural labor force to cities. Over the past few years, there has been a notable increase in both per capita disposable income and consumption expenditure among rural residents. This upward trend has directly contributed to significant enhancements in the living standards of left-behind children. Whether the change of economic status is an important social factor affecting changes of self-esteem of left-behind children remains to be explored. Therefore, we selected two social indicators of per capita disposable income and per capita consumption expenditure of rural residents to investigate the influence of economic conditions on the self-esteem. Secondly, educational level is also closely related to psychological changes. Research findings have indicated a positive correlation between the level of education and self-esteem, suggesting that individuals with higher educational attainment tend to exhibit higher levels of self-esteem [23]. In recent years, there has been a growing focus from the government on the education of left-behind children, accompanied by increased investments in educational resources. Meanwhile, the widespread implementation of nine-year compulsory education has contributed to an upward trajectory in the educational attainment of left-behind children. Therefore, we selected two social indicators, namely educational expenditure, junior high school enrollment rate, to investigate the influence of educational levels on the self-esteem.

To sum up, in view of the fact that large-scale and continuous longitudinal studies on the self-esteem of left-behind children are rare, cross-temporal meta-analysis can form large-span time studies by combing empirical studies at different time points, so as to examine the age effect of research variables and explore the development laws behind them. Therefore, we employ cross-temporal meta-analysis to investigate the evolving pattern of self-esteem over the years, and adopts the lag analysis to explore the connection between social indicators and the self-esteem of left-behind children.

Method

Research tools

Self-esteem scale (SES): SES was compiled by Rosenberg [7] in 1965 to evaluate participants’ overall self-esteem. There are 10 items in the scale, with half of the forward and reverse items. Each item is scored on 4 points, with the total score between 10 and 40. Additionally, participants were divided into three groups in accordance with their scores: those with scores of 25 or less were labeled as having low self-esteem, those with scores ranging from 26 to 32 were classified as having moderate self-esteem, and those with scores of 33 or more were classified as having high self-esteem.

Literature collection

Literature collection criteria

Considering the specific circumstances of left-behind children, we adopted the following criteria for literature collection and screening: (a) the study focused on children under 18 who were left behind; (b) the study used the Rosenberg Self-esteem Scale; (c) the research results provided explicit self-esteem data, such as the sample size and statistical measures like the mean and standard deviation, both for the entire sample and any applicable sub-samples; (d) If multiple studies were published by the same authors using the same batch of data, only the one with the most complete data and the earliest publication date should be selected.

Literature collection results

According to the above standards, we conducted full-text search with the keywords of “self-esteem”, “left-behind” and “left-behind children” in the databases of CNKI, Wan fang Database, WHIP Journal, ProQuest and Web of Science. Finally, a total of 72 documents meet the requirements of cross-temporal meta-analysis. They were published between January 2008 and December 2022, and the documents were distributed in each year. The PRISMA flowchart is displayed in Fig. 1.

Fig. 1

PRISMA flowchart

Moreover, for the study to maintain precision, the term “year” in this context represents the time of data collection. If the literature explicitly stated the year of data collection, that particular year was utilized. If it was not mentioned in the literature, it can be obtained by subtracting two years from the publication date [24]. For example, for the literature published in 2008, we believe the year of collection is 2006. Therefore, the study covers the years from 2006 to 2020, with a total sample size of 29,835 left-behind children (Table 1).

Table 1 Literatures and sample distribution included in the study

Social indicators

As mentioned in the introduction section, previous studies have considered social factors such as economic conditions and educational level as influencing factors on self-esteem.

Moreover, a cross-temporal meta-analysis of the specific components of self-esteem, namely physical self-esteem, employed economic conditions and educational level [25]. Additionally, it is noted that left-behind children are in primary and secondary education stages. Therefore, based on previous studies and the subjects and focus of the current study, four social indicators were selected from the perspectives of economic conditions and education level, including per capita disposable income of rural residents, per capita consumption expenditure of rural residents, educational expenditure and junior high school enrollment rate. All the aforementioned statistics come from the China Statistical Yearbook, which released by the National Bureau of Statistics in 2022 [26].

Results

Group differences in self-esteem of left-behind children

To comprehensively understand the self-esteem of left-behind children, we computed the mean and standard deviation of self-esteem. Subsequently, we conducted a variance analysis of self-esteem across various educational stages and regions (For detailed information, please refer to the supplementary materials). The results showed that the average self-esteem of left behind children was 27.51. In addition, the results of the analysis of variance indicated that there were no significant differences (F (3, 66) = 1.30, P > 0.05) in self-esteem among left-behind children across various educational stages. Similarly, no significant differences (F (3, 64) = 0.53, P > 0.05) in self-esteem were observed among left-behind children from different regions.

Overall change in self-esteem of left-behind children with the years

In order to intuitively reflect the changes in the self-esteem level over time, we drew a scatter plot with the year of data collection as independent variable and the total score of self-esteem as dependent variable. Left-behind children score progressively higher on self-esteem between 2006 and 2020 (see Fig. 2).

Fig. 2

Changes in self-esteem score of left-behind children, 2006–2020

In order to quantify the trend of change more accurately, we explored the correlation between self-esteem scores and year of data collection. The results indicated that controlling for sample size, the correlation coefficient between self-esteem score and year was 0.33, P < 0.01. To further examine the temporal effects of changes in self-esteem, we conducted a regression analysis, using self-esteem score as the dependent variable and year as the independent variable. The results indicated that after weighting the sample size, the year was found to significantly predict the self-esteem (β = 0.25, P < 0.05). In addition, after controlling for variables such as educational stage, region (participant source), the regression coefficient (β = 0.28, P < 0.05) remained significant.

The extent to which the self-esteem of left-behind children changes

The aforementioned findings indicated an increase in the self-esteem. However, it was challenging to determine the exact magnitude of this increase based solely on these results. To clarify the range of changes in self-esteem, we established a regression equation (y = 0.14x–250.10, x is the year of data collection, y is the self-esteem score) under the premise of weighting the sample size to predict the mean scores of the first and last year of included studies. Then, 2006 and 2020 were put into the regression equation respectively to obtain scores of these two years and find the difference between them (Table 2). Finally, we evaluate the authenticity of the difference by calculating the effect size d (formula 1).

$$d = (M_{2020} - M_{2006}) / M_\text{SD}$$

(1)

Note

M_SD is the mean standard deviation obtained by averaging the standard deviations of all included studies.

Table 2 The change of mean self-esteem scores from 2006 to 2020

As shown in Table 2, the mean score increased from 30.74 in 2006 to 32.7 in 2020. The mean standard deviation was 3.74, and the effect size was 0.52. According to Cohen’s [27] guidelines, the effect size between 0.20 and 0.50 is a small effect, between 0.50 and 0.80 is a moderate effect, and above 0.80 is a large effect. The effect size of self-esteem change in this study is 0.52, which is a moderate effect, indicating that the self-esteem level has increased significantly in the past 15 years.

Correlations of left-behind children’s self-esteem with social indicators

Although the self-esteem of left-behind children has increased over the years, we have not yet determined what caused it. As mentioned earlier, cross-temporal meta-analysis can not only explore the changing trend of psychological variables over time, but also reveal the impact of social changes on psychological development through the analysis of the association between psychological variables and social indicators. In other words, we can select social indicators that are closely related to psychological variables of a group to conduct correlation or lag correlation analysis with the scores of psychological variables. For example, if there are significant correlations between psychological variables and social indicators 5 years before data collection, 3 years before data collection and the year in which data was collected, it can be shown that social factors have an impact on psychological development. Therefore, we conducted the correlation between self-esteem scores of left-behind children and social indicators of 5 years prior, 3 years prior and the current year respectively (see Table 3), namely, the correlation between self-esteem scores of 2006–2020 and social indicators of 2001–2015 and 2003–2017.

Table 3 Correlations between social indicators and self-esteem score of left-behind children

The results are presented in Table 3. After controlling the sample size, all correlations were significant. Among them, correlations between social indicators of the year and self-esteem scores were relatively stronger. It can be seen that economic conditions (per capita disposable income and per capita consumption expenditure of rural residents) and educational level (educational expenditure, junior high school enrollment rate) may be important factors affecting the self-esteem of left-behind children.

Discussion

The self-esteem level of left-behind children has steadily improved

By employing cross-temporal meta-analysis, this study conducted an analysis of 72 documents from all over the country, investigating the evolving tendency of self-esteem among left-behind children from 2006 to 2020. In the past, cross-temporal meta-analyses on self-esteem have primarily concentrated on college and middle school students [28,29,30], with little attention given to the subgroup of left-behind children. Left behind children are a special group that has emerged in the process of urbanization in China, and they account for a large proportion of Chinese youth. Exploring the changes in self-esteem of left behind children is of great significance for safeguarding their mental health and promoting social harmony, considering their unique status in the context of China’s urbanization.

Through the analysis, the results showed that the self-esteem increased from low to moderate level. The results are consistent with the findings of Li [31] regarding the changes in self-esteem among middle school students but diverge from those reported by Liu [32]. One potential explanation for this disparity may be attributed to the differing timeframes in which the studies focused on the changes in self-esteem among adolescents. Liu and colleagues conducted their research spanning from 1996 to 2009, while our study investigated the self-esteem changes among left-behind children from 2006 to 2020 (date of data collection), encompassing a broader range of years. During these years, there has been significant improvement in the societal environment. The heightened attention from the government towards the well-being of left-behind children, along with the implementation of Opinions on Strengthening the Care and Protection of Rural Left-behind Children have played a positive role in improving the self-esteem.

Correlation between social indicators and self-esteem of left-behind children

Consistent with social development, left-behind children’s self-esteem is getting stronger and stronger. This result confirmed our prediction that self-esteem has increased over the past 15 years and suggested that the impact of social environment on individuals may exceed genetics and family [19]. In the years with higher disposable income, higher consumption expenditure, higher educational expenditure and higher junior high school enrollment rate, left-behind children tend to have higher self-esteem.

First of all, there was a noteworthy correlation between economic conditions and self-esteem of left-behind children. The per capita disposable income and per capita consumption expenditure of rural residents reflect the level of economic development in rural areas. The findings revealed that the economic development and consumption capacity improvement in rural areas contributed to the development of self-esteem. Left-behind children are in a sensitive stage of physical and mental advancement [33]. In the process of growth, it is inevitable to compare themselves with other peers in material aspects [34]. Low family economic status tends to make left-behind children feel inferior in comparison with their peers, which is not conducive to the development of self-esteem [35]. In recent years, with the government’s support for migrant workers in cities and the promotion of rural revitalization, the income level of rural residents has improved significantly. Previous studies have shown that family monthly income significantly impacts the peer relationships of left-behind children. A higher family monthly income is correlated with greater acceptance and respect in peer interactions among left-behind children, resulting in heightened levels of self-esteem [36, 37]. In addition, with the increase of income level, the quality of life of left behind children has also significantly improved, which can meet their own material needs., Moreover, the care of all sectors of society for “left-behind children” makes them more confident in peer communication and comparison, thus improving their self-esteem [38].

Secondly, the educational level plays a crucial role in the increased self-esteem observed in left-behind children. Educational expenditure pertains to the actual funds allocated for education within the budgets of both central and local financial departments, and junior high school enrollment rate refers to the proportion of junior high school graduates entering higher education. The findings from correlation analysis indicated that education can subtly affect the cognition, and the improvement of education input and education level are beneficial for fostering the development of self-esteem. In recent years, with the national emphasis on education and the widespread implementation of nine-year compulsory education, the educational environment and the level of education among left-behind children have witnessed significant improvements. A good educational environment provides a good psychological and physical development environment for them. Studies have found that the learning environment has a direct positive impact on learning attitudes and outcomes. A conducive learning environment is associated with improved academic performance, which, in turn, plays a crucial role in enhancing self-esteem [39]. In addition, the study also found that the improvement of education level can lead individuals to have higher self-evaluation of themselves, promoting the improvement of self-esteem level [40]. Therefore, as the education level of left behind children improves, their evaluation of themselves becomes more positive.

Research deficiencies and prospects

Despite the aforementioned positive outcomes, this study does have certain limitations. First, this study adopted Rosenberg10-items self-esteem scale, but Hao [41] found that the reliability and validity of item 8 in the self-esteem scale were not high among Chinese students, so some studies used 9-items self-esteem scale excluding item 8. In the future, we can conduct a cross-temporal meta-analysis of the literature using 9-items self-esteem scale and compare it with the results of this study. Secondly, this study analyzed the relationship between social indicators and self-esteem, but did not further verify the underlying impact mechanism. In the future, we can further explore the internal mechanism of the impact of social factors on self-esteem.

Conclusion and suggestion

This study employed cross-temporal meta-analysis to delve into the self-esteem of left-behind children (2006–2020). The outcomes demonstrated that: (1) The level of self-esteem of left-behind children has been rising between 2006 and 2020; (2) Social indicators (economic conditions and educational level) are significantly correlated with self-esteem of left-behind children.

The study’s findings offer guidance on improving the self-esteem and mental health of left-behind children through a multifaceted approach. Firstly, Financial support and social assistance programs should be implemented to elevate the average disposable income of rural residents, thereby alleviating economic burdens on families. Specific aid plans, such as scholarships and living allowances, should be tailored for left-behind children to assist them in meeting their basic needs. Secondly, addressing educational disparities is crucial. Increasing investment in rural education resources ensures that left-behind children have access to a conducive learning environment and qualified teachers. Incentivizing educators to work in rural areas and implementing supportive policies are essential steps. Thirdly, besides the economic and educational support provided by the government, the efforts of communities and families are also indispensable to enhance the self-esteem of left-behind children. Establishing community mental health support centers and encouraging social connections among children in their neighborhoods address their sense of isolation. Simultaneously, fostering improved communication within families through targeted training programs creates a positive and supportive environment for the children’s emotional well-being. These recommendations, when collaboratively implemented at the governmental, community, and family levels, will ensure comprehensive support for the mental health and self-esteem of left-behind children.