The power of family: The impact of family culture on marriage stability in China

Abstract

Family culture is deeply rooted in the transformation and development of Chinese society and has a profound influence on individual behavior and family decision-making. This paper examines the impact of the family culture of China on marriage stability. By using data from the China Family Panel Studies, this paper shows that family culture is significantly positively associated with marriage stability. The estimation of the instrumental variable—the number of memorial archways (paifang) per 100,000 persons in each province—and a series of robustness checks all support the major results. It is also found that the positive impact of family culture on marriage stability is mainly seen for people with weaker perceived importance of family well-being, for those with lower levels of education, and for individuals living in a small family. Finally, the positive effect of family culture on marriage stability is primarily observed in older generations, specifically those born before 1969.

Keywords

Family culture marriage stability China

Introduction

Several changes in the family have been seen during the past half-century, such as the retreat from marriage and rising rates of divorce, cohabitation, and remarriage, but no change has altered United States family life as deeply as the fact that most marriages no longer last a lifetime (Bumpass, 1990). In the last few years, the rising divorce rate has also become one of the most prominent features of the Chinese family. According to data published by the National Bureau of Statistics, divorce rates rose rapidly from 1.53‰ in 1978 to 10.37‰ in 2018, the crude divorce rate also increased from 0.3‰ in 1978 to 2.79‰ in 2015.¹ However, the crude divorce rate is susceptible to the influence of population structure, age at first marriage, and marriage rates across cohorts, therefore it is not a reliable indicator of trends in marriage stability. Yu et al. (2020) recently conducted a cross-country comparison of divorce rates among different gender groups and age groups in first marriage cohorts, showing that compared with Western developed countries, the proportion of divorced people and the divorce rate within 10 years of marriage in China are both at very low levels. The divorce rate of couples who entered into marriage between 2000 and 2004 is only 3.7% within 10 years of marriage, which is far lower than the rate of about 20% in Europe and the United States. This means that although marital stability has declined in recent years, divorce is still not a common phenomenon in China. What is the reason for this? The answer to this question is important for understanding the changing nature of marriage and family in China. This study will offer a cultural explanation for the lower divorce rate in China.

Culture can remain stable for a long time (Hitlin and Piliavin, 2004; Parsons and Shils, 1951). Many researchers have suggested that culture may have a significant impact on people's behaviors, such as divorce (Guiso et al., 2006). Since the second demographic transition, American culture has become more individualistic (Amato and Irving, 2006). With the “me first” generations beginning to pursue personal freedom and happiness, traditional marriage obligations have gradually eroded. This could partly explain the rising divorce rate in the United States after 1960. Meanwhile, some scholars have also found a relationship between specific social cultures and marriage dissolution. For example, Ghanaians with matrilineal kinship ties have a high divorce rate (Oppong, 1974; Takyi and Gyimah, 2007) because married couples rarely put their resources together—instead they give priority to their matrilineal kin (Clark, 1999). This practice of being loyal to one's lineage weakens the conjugal bond. Moreover, divorced women may more easily reintegrate into their matrilineage and get assistance from family members. Immigration has also been used to disentangle the impact of culture on marriage disruption. Eeckhaut et al. (2011) found that “migration marriages” among otherwise ethnically homogamous partners result in higher rates of divorce than among ethnically homogamous marriages between partners from the same country, as migration brings cultural differences which may cause tensions within the marriage. Furtado et al. (2013) used home-country divorce rates as a proxy of culture to explore the impact of culture on divorce. They found that immigrants coming from countries with low divorce rates tend to be less likely to divorce if they are living within their own ethic groups, and divorce culture has a stronger impact on females’ divorce decisions than males’.

This study contributes to an emerging literature that examines the role of culture in determining behavior. For example, Guiso et al. (2006) showed that culture can exert a broad impact on economic outcomes by influencing individuals’ preferences or values. Fernández and Fogli (2009) also used an epidemiological approach to test the causal relationship between culture and a series of household decisions such as fertility, migration, and labor supply. Following Fernández (2011), Furtado et al. (2013) used the same approach to test the impact of local divorce culture on marriage disruption. Furthermore, several Chinese scholars have begun to research the effect of the unique Chinese culture on socioeconomic outcomes. Zhang (2019), also using China Family Panel Studies (CFPS) data, found that clan culture is negatively related to enrollment in China’s state pension program because it is positively related to the value that having a son can offer financial support when people enter old age. Hu and Li (2020) also found a positive link between ancestor worship and household eldercare. Chen et al. (2019) showed that parents living in regions with stronger Confucian norms tend to save less as the probability of receiving old-age support from their children is higher than in other regions. In addition to filial piety, “pacifist” doctrines promoted by Confucian norms are capable of reducing conflicts and peasant rebellions (Kung and Ma, 2014). This study adds to this body of knowledge by examining the role of China’s family culture on marriage and the findings will enhance our understanding of social norms and the socio-cultural forces of maintaining family harmony.

Meanwhile, this study also contributes to a growing literature on the causes of marriage stability. Many scholars have attempted to explore the reason for marriage dissolution from different perspectives. On the one hand, some researchers have attributed it to macro-level structural causes, including legal factors (Parkman, 2000; Wolfers, 2006), economic factors such as unemployment (Jensen and Smith, 1990), and cultural factors (Amato and Irving, 2006; Furtado et al., 2013). On the other hand, some micro-level factors such as the age of marriage (Thornton and Young-DeMarco, 2001), race (Heaton, 2002), religion (Bramlett and Mosher, 2002), income (Rogers, 2004), living with divorced parents (Wolfinger, 2000), premarital cohabitation (Axinn and Thornton, 1992; Y Zhang, 2017), the increase of women's education and status (Amato, 2010), and migration (Li, 2018), and some personality traits also can predict marriage dissolution. But these factors cannot fully account for the lower divorce rate in China when compared to Western countries. In this study, I will offer a unique cultural explanation for marriage stability and present new evidence from China.

In China, family culture exerts an impact on the family system and household structure by constructing social norms and order, morals, and customs, which can help limit behavior (Wang, 2008). Therefore, it plays an important role in maintaining marriage stability (Yu et al., 2020). This study will examine the causal relationship between family culture and marriage dissolution by using CFPS data. The results suggest that China’s family culture is significantly negatively related to marriage dissolution and can help promote marriage stability. This result shows considerable robustness after multiple robustness tests, such as controlling for personal, family, village, and province characteristics, replacing the core variables, replacing the model, and overcoming endogeneity problems. Furthermore, it is found that the positive effect of family culture on marriage stability is predominant among groups with weaker perceptions of the importance of family well-being and lower levels of education. The study also found that the positive effect of family culture on marriage stability decreases with an increase in family size. Finally, the positive effect of family culture on marriage stability is primarily observed in older generations, specifically those born before 1969.

The remaining part of this article proceeds as follows: the second section provides a research background; the third section describes the data and the construction of variables; the fourth section introduces the empirical strategy; the fifth section examines the baseline results and robustness checks; the sixth section discusses further research about the moderation role of perceived importance of family well-being, education, and family size; and the last section provides a brief conclusion to this study.

Background

Confucian culture, which advocates the virtues of harmony and subordination, plays an important role in maintaining social stability and regulating social order. It takes “benevolence” and “propriety” as its core values to regulate people's behavior and guide them to avoid conflicts and violence. Therefore, Confucianism was promoted by ancient China's imperial rulers as a state orthodoxy and ideology for around 2000 years.

The family is the most basic unit of traditional Chinese society, so “familyism” is the core of Confucianism (Fukuyama, 1999). The most salient characteristics of the traditional Chinese family were patriarchy, the patrilineal system, and patrilocal residence. The Chinese patriarchal structure consists of two axes: the generational axis, which means elders have prestige and power over the young generation; and the gender axis, which means the female will always be subordinate to the male (Santos and Harrell, 2017). Therefore, Chinese family culture is actually a male-centered differential ethical structure constructed based on the closeness of blood relations.

Ancestor worship, the most important and the oldest tradition for Chinese, becomes a source of family strength and gives strong support to patriarchal power (Weber, 1951). Its main manifestation is in the visiting of gravesites for deceased family members as a family unit during the Qingming (“Tomb-Sweeping”) Festival. Ancestor worship has the following functions. First, it is a collective activity that helps to strengthen family ties and kinship attachments, enhance family unity, and demonstrate the strength of the family (Hsu, 1971). Second, this practice can express good wishes and prayers for family members (Freedman, 1966). Third, it can uphold the patriarchal status and authority and reassert the hierarchical status and boundaries of the family (Cohen, 1990).

Marriage is a key bond for maintaining blood ties. The influence of family on marriage is reflected in the choice of a spouse and the maintenance of the relationship after marriage. For example, in traditional Chinese society, when people needed to choose a spouse, they had to obey “the command of parents, [and] the words of a match-maker”. Assortative matching is homogamy (as expressed in the traditional concept of mendang hudui, meaning roughly “family doors of equal size”), which means that couples are well-matched in terms of their social status. The principle of post-marital relations was summed up in the “three obediences and four virtues”. The three obediences were that women need to obey their father before marriage, become subordinate to their husband after marriage, and raise their children after their husband's death. The four virtues were fidelity, physical charm, propriety in speech, and efficiency in needlework. The traditional family culture was also strict with women. A woman did not have the right to divorce freely. On the contrary, a woman was subject to unconditional repudiation by her husband if one of seven conditions (childlessness, adulterousness, disobedience to one's husband's parents, gossiping, thieving, jealousy, or sickness) was present. Women had to preserve loyalty and fidelity to their husband. A widow who did not remarry or even committed suicide after her husband's death would be considered and glorified as a chaste woman or lienü. Although the traditional family culture was more about defending the interests of men, it did not completely deny the obligations of men. For example, the family culture still put forward the value coordinates of “righteousness of husband and obedience of wife” in marriage.

In sum, Chinese family culture affects marital stability mainly in the following ways. First, family harmony is considered the core of the family culture and the foundation of family prosperity. The value of pacifism advocated by the Chinese family culture emphasizes the importance of harmony and stability between husband and wife, guiding family members to love their own families and reducing family violence and conflict. Secondly, subordination is another principle of Chinese family culture. The differential patriarchal structure requires family members’ loyalty to the male members of the family and to the senior generation. Moreover, in traditional Chinese society, due to the practice of patrilocal residence, if the relationship broke down, a woman may have lost her basic security and the economic basis for her survival. For reasons of economic rationality, women may have been more careful in maintaining the marital relationship. Thirdly, divorce is stigmatized in traditional family culture and is even considered unfilial—rejected and condemned by families and village acquaintances, it can bring disgrace to parents and the family in general. This potential public pressure also inhibits divorce to some extent. Fourth, solidarity is the final characteristic of family culture. Marriage in China is not just about the husband and wife, it also involves the parents and blood relatives of both parties (Hsu, 1981). Thus, it is a public matter within the family (Fei et al., 1975). When a marriage is in crisis, the elders and relatives of the family also mediate between the parties. Out of respect for the opinions of family elders, couples will consider and reconsider marriage decisions carefully, in an attempt to prevent impulsive divorce.

Methodology

CFPS survey data and other data sources

The main data used in this study come from the baseline survey data of CFPS, conducted by the Chinese Center for Social Science Survey (ISSS) at Peking University. These data include five main types of questionnaires: communities, households, household members, adults, and children.

The longitudinal surveys began in 2010 and were conducted every two years. So far, five waves have been completed. The baseline survey in 2010 covers 94.5% of the total population of the Chinese mainland, that is, approximately 14,798 households in 25 provinces/municipalities/autonomous regions, excluding Hong Kong, Macao, Taiwan, Tibet, Xinjiang, Qinghai, Ningxia, Hainan, and Inner Mongolia. Therefore, the data are nationally representative. In terms of the sampling methodology, the CFPS uses probability-proportional-to-size sampling with implicit stratification. The samples were obtained in three stages: the first-stage sampling unit was either an administrative district in urban areas or a county in rural areas; the second-stage unit was either a neighborhood community in urban areas or an administrative village in rural areas; and the final-stage sampling unit was the household.

The CFPS 2010 baseline survey distributed samples to 19,986 households in 25 provinces, 162 administrative districts/counties, and 649 communities/villages. Finally, the baseline survey completed interviews with 14,960 households in 635 villages/residences and identified 57,155 eligible family members. About 42,590 of those family members (33,600 adults and 8990 children) completed the individual questionnaire. After excluding the observations with missing values in the main variables, this study finally obtained 23,041 records.

Although the CFPS has conducted five completed waves as of 2021, I primarily use baseline survey data from the 2010 survey. As this study focuses on the role of family culture, CFPS 2010 is the only dataset that provides measures of family culture, such as whether or not family members visit the gravesites of family ancestors, whether or not the family has a genealogical tree, and the proportion of the most common family names in the community. The following waves in 2012, 2014, 2016, and 2018 no longer included such questions. For this reason, this study uses only the baseline survey data.

In addition to the CFPS data, I also utilize a database of China’s existing memorial archways (paifang), based on material collected by three Chinese scholars (Zhao et al., 2016). The work of collecting data was divided into three steps. The first step was to widely collect memorial archway-related information in China, such as the name of memorial archway, geographic location, construction year, shape, material, and cultural relic protection status from the Internet by using search engines, from which was then formed a database. In the second step, this information was substantiated, corrected, filtered, and categorized according to China's Memorial Archways edited by Jin (2003) and other authoritative monographs. The final step was to supplement and enrich the memorial archway data by searching for the latest relevant literature in the CNKI (https://www-cnki-net-443.web.bisu.edu.cn/), VIP (http://www.cqvip.com/) and Elsevier (http://www.elsevier.com/) databases, and to guarantee the integrity and accuracy of the memorial archways dataset.

Until the deadline for data collection (at the end of 2013), the study collected a total of 1045 old memorial archways, divided into seven types:² jiexiao fang,³ gongde fang,⁴ baisui fang,⁵ biaozhi fang,⁶ miaoyu fang,⁷ lingmu fang,⁸ and zongzu fang.⁹ Considering that biaozhi fang and lingmu fang were built to divide geographic space and commemorate the dead, respectively, the five remaining types of memorial archways all reflect the family culture of loyalty, filial piety, and chastity. Therefore, I only use these five types of memorial archways for a total of 802 in my study.

Main variables

Dependent variable

According to the variable of marital status in the CFPS 2010 individual questionnaire, the dependent variable “marriage stability” was operationalized as “whether the respondent is divorced” (1 = yes, 0 = no) (De Paola and Gioia, 2017; Dzara, 2010). Unmarried, cohabitating, and widowed people are discounted from this study.

Independent variable

The household questionnaire of the CFPS 2010 can provide us with the key information regarding family culture. Therefore, family culture was measured by “whether or not the family visits gravesites for family ancestors regularly? (1 = yes; 0 = no)”. Ancestor worship is the main embodiment of family culture. The ritual of sweeping the ancestors’ graves on the annual Qingming Festival is an internal family affair, limited to blood relatives. It gathers scattered family members together to achieve the identity of the consanguineous family and acquire the sense of belonging to the family community. It is also an important manifestation of filial piety and patriarchy that are core aspects of Chinese family culture. Therefore, visiting gravesites is the main symbol of family culture and I will use this variable as the main proxy for family culture.

Beyond that, to obtain the stable measurement of the impact of family culture on marriage stability, I also use another two variables “whether or not the family has a genealogy” (1 = yes; 0 = no) and “daxing zhanbi” (“the proportion of the most common family names in a community”) as a robust check. A family genealogy records the origin, evolution, branches, migration, and reproduction history of each consanguineous family, as well as the unique customs, etiquette, and the deeds of important family members; as such, it is an important material vehicle of family culture (Greif and Tabellini, 2017; Peng, 2004; Tsai, 2007). Secondly, in China, the higher the proportion of the most common family name in a village, the closer the local clan relationship and the stronger the family culture (Chen and Chen, 2018). Although the latter variable is used to measure clan culture generally, family culture is an important part of clan culture. Therefore, these two variables are also considered as suitable measurements of family culture.

Finally, according to the visiting of gravesites at the household level, I calculated the average level of ancestor worship in each village as a measure of community-level family culture.

Control variables

Several control variables at the individual level, family level, and community level are also considered. I first control for individual characteristics, including gender (1 = male, 0 = female), age, ethnicity (Han = 1, other minority ethnicity = 0), employment (1 = yes, 0 = no), education (years of formal education), self-reported health status (1–5 ordered variable from very unhealthy to very healthy), Communist Party of China (CPC) membership (1 = yes, 0 = no), and hukou status (1 = urban, 0 = rural). Furthermore, household-level characteristics are also included, such as the number of children, as well as the number of siblings and the log-transformed income per capita in the household. Lastly, I also control for type of residence (1 = urban, 0 = rural) as the community characteristic.

Empirical strategy

Basic modeling

To examine whether family culture can improve marriage stability, I estimate the following Probit regression equation:

Y_{i j} = β_{1} f a m i l y c u l t u r e_{i j} + β_{2} X_{i j} + β_{3} D_{i j} + β_{4} C_{i j} + ε_{i j}

(1)

where Y_ij is the marriage status for respondent i in village j. Family culture is the core variable I am interested in, which is operationalized as whether or not the family visits its gravesite regularly. X_ij is a vector of individual characteristics, including gender, age, ethnicity, work, years of formal education, self-reported health status, CPC membership, and hukou status. D_ij is a set of household-level covariates, including the number of children, number of siblings, and the log-transformed income per capita of the household. C_ij is a community-level control, including the type of residence. ε_ij is a disturbance term. I also use robust standard errors that are clustered at the province level.

Equation (1) is first estimated using Probit regression and the goal of this study is to identify β₁_. However, this result may still be challenged by the problem of endogeneity. Generally, there are three main sources of endogeneity: reverse causality, omitted variables, and measurement error. First, family culture is an intrinsic, inherent product of a family history that does not change with an individual's behavior. That is to say, visiting one's ancestors’ gravesite is a fixed ritual and it will not stop because of personal marriage disruption. Therefore, the likelihood of estimation bias caused by reverse causation is very low. Second, because family culture is complex and abstract, there is the possibility of measurement error in the measurement of this variable. Therefore, I also use the presence/absence of a family genealogy, the proportion of most people having the identical family name in a community, and community-level ancestor worship as robust checks to mitigate the possibility of measurement error. Third, as for the missing error problem, I have controlled for individual, household, and community characteristics which may influence marriage disruption, but omitted variables could still be a cause of biased regression estimates.

Considering this, the ideal way to investigate the impact of family culture on marriage disruption is probably to adopt the instrumental variable (IV) approach. I will discuss next whether this potential instrument is a strong predictor for family culture below.

IV approach

Although I control for the above variables, some omitted variables may cause endogeneity problems. To purge a possible bias that might stem from omitted variables, I will use the IV method. I constructed an IV of the memorial archways (paifang) per 100,000 persons in each province.¹⁰ The number of memorial archways is obtained from a study (Zhao et al., 2016), which collects the locations, types, and numbers of memorial archways in Chinese mainland—the population in each province (during the Xuantong period of the Qing Dynasty) comes from a study of a population survey (Hou, 1998).

The number of memorial archways per 100,000 persons in each province is closely related to the intensity of local family culture. In ancient China, memorial archways were originally built to pray for a good harvest, but they were gradually afforded political, cultural, and moral functions by the ruling class. The memorial archway was a typical materialization of Confucian culture. In the Ming and Qing dynasties, Confucianism had become deeply ingrained in Chinese society and its new form, the Cheng–Zhu school of neo-Confucianism (Cheng Zhu lixue), had become the main orthodox ideology by then. Memorial archways were gradually used as an important tool to promote Confucian ethics of subordination and loyalty (zhongxiao jieyi) in the family and other laudable behavior. For example, “zhenjie” memorial archways were established to honor “chaste women”: widows who had sworn never to remarry and those who committed suicide after their husband's death. They represented the wife's purity, loyalty, and fidelity to their husband and the deeds of chaste women considered as virtuous, and a code of conduct that the bureaucracy was eager to promote.¹¹ Therefore, the more memorial archways in the region, the greater the impact of Confucian culture in that area. As an important part of Confucian culture, the strength of family culture should be in line with the wider impact of Confucian culture.

On the other hand, the number of memorial archways is exogenous and does not have a direct effect on micro-level and individual marriage decisions and it can only influence individual marriage behavior by exerting an impact on the present family culture. Thus, this variable is a valid instrument since it satisfies the condition of exogeneity and correlation. I then match the provinces in CFPS 2010 with the numbers of memorial archways per 100,000 persons in each province and subsequently obtain a variable of how many memorial archways there are in each province.

Results

In this section, I will present the descriptive results first and estimate Equation (1) using Probit and the IV approach. The robust standard errors are clustered at the province level.

Descriptive results

Table 1 presents the descriptive statistics of the dependent, independent, and other control variables. I have deleted 3492 samples who were unmarried, cohabitating and widowed, and restricted the sample to married or divorced people. After cleaning the important variables, a total of 1.6% of respondents is divorced in the whole sample. For those households with a strong family culture, the percentage of divorced couples is below average, at only 1.38%. As for households lacking strong family culture, the percentage of divorced couples is much higher than the average, accounting for 2.17%. Furthermore, family happiness is greater and family size is larger for those individuals in households with family culture.

Table 1.

Descriptive statistics of selected variables.

Variable	Mean/percentage	Standard deviation	Family culture
Variable	Mean/percentage	Standard deviation	Yes	No
Marriage stability (divorce)	1.6%	0.126	1.38%	2.17%
Family culture	71.16%	0.453	—	—
Gender (male)	49.09%	0.500	49.01%	49.27%
Age	47.574	13.300	47.41	47.97
Ethnicity (Han)	92.07%	0.270	92.92%	89.98%
Work (yes)	54.77%	0.498	57.30%	48.52%
Education	6.064	4.841	6.15	5.82
Health (1–5)	4.139	1.027	4.14	4.12
Communist Party of China membership	8.54%	0.280	8.69	8.17
Hukou status (urban)	29.58%	0.456	28.67%	31.83%
Number of children	1.943	1.147	1.93	1.96
Number of siblings	3.077	1.888	3.09	3.02
Log-transformed income per capita	8.652	1.064	8.67	8.59
The type of residence (urban)	45.89%	0.498	44.91%	48.31%
Number of memorial archways	23.780	22.949	24.35	22.36
Family happiness (1–5)	4.623	0.681	4.63	4.59
Family size (number of family members)	4.269	1.802	4.31	4.18

Note: Data are from China Family Panel Studies 2010.

To better visualize geographic differences in family cultural intensity, Figure 1 presents the percentage of ancestor worship in each province in CFPS.

Figure 1.

The percentage of ancestor worship in each province in the China Family Panel Studies.

Baseline results

The baseline results are reported in Table 2. As shown in columns (1) to (4), the members of families advocating ancestor worship beliefs show a significantly lower probability of marriage dissolution. This means family culture may play an important role in maintaining marriage. The results are robust after including individual, household, and community control variables. The probability of divorce for the members of families with strong family culture is significantly reduced, by 0.7%, according to the margin effect.

Table 2.

The impact of family culture on marriage stability.

	(1)	(2)	(3)	(4)
Family culture	−0.183***	−0.183***	−0.177***	−0.173***
	(0.043)	(0.045)	(0.046)	(0.059)
Gender		0.181***	0.156***	0.162**
		(0.045)	(0.046)	(0.063)
Age		0.048***	0.059***	0.057***
		(0.012)	(0.013)	(0.013)
Age squared		−0.0005***	−0.0005***	−0.0005***
		(0.000)	(0.000)	(0.000)
Ethnicity (Han)		−0.087	−0.143*	−0.159
		(0.080)	(0.084)	(0.102)
Employment (yes)		−0.028	−0.021	−0.013
		(0.047)	(0.048)	(0.038)
Education		0.015***	0.009	0.007
		(0.006)	(0.006)	(0.008)
Health		−0.032	−0.025	−0.027
		(0.022)	(0.023)	(0.032)
Communist Party of China membership		−0.270***	−0.247***	−0.244***
		(0.083)	(0.086)	(0.075)
Hukou status (urban)		0.425***	0.291***	0.222***
		(0.051)	(0.056)	(0.061)
Number of children			−0.416***	−0.409***
			(0.033)	(0.061)
Number of siblings			−0.015	−0.014
			(0.013)	(0.014)
Log-transformed income per capita			−0.055**	−0.064**
			(0.024)	(0.026)
The type of residence (urban)				0.158**
				(0.066)
Intercept	−2.020***	−3.160***	−2.438***	−2.363***
	(0.034)	(0.312)	(0.363)	(0.297)
Pseudo R-squared	0.005	0.050	0.100	0.102
Chi-squared	17.585	188.992	378.623	1012.967
Number of observations	23,041	23,041	23,041	23,041

Notes: standard errors in parentheses. Robust standard errors in column (4) are clustered at the province level. * p < 0.1, ** p < 0.05, *** p < 0.01.

With regard to the control variables, males show higher divorce risks than females. The relationship between age and the probability of divorce is non-linear, showing an inverted U-shaped relationship. People with higher education show higher divorce risks in column (2), although this becomes insignificant when the model includes the household and community control variables. CPC members show a lower probability of marriage disruption. The divorce risk is higher for those respondents with urban hukou than that for those with rural hukou. Moreover, children in the family will increase the cost of divorce, so the results indicate a significant negative correlation between the number of children and marriage disruption. Higher household income per capita is also negatively associated with marriage dissolution. Finally, people living in urban areas show a higher risk of divorce than those living in rural areas.

To mitigate the attenuation bias due to omitted variable problems, which are unable to be resolved by the limited control variables, I adopt the IV approach in the next subsection.

IV results

To guarantee that the IV is not directly related to divorce, this paper adds to the existing controls the additional province-level control variables: “number of libraries”, “number of hospitals”,¹² “gross domestic product per capita”,¹³ and “average education level”.¹⁴

Column (1) in Table 3 reports selected results from the first stage. It indicates that the IV, the number of memorial archways per 100,000 persons, is highly significantly related to family culture. According to the meaning represented by the memorial archways, an area with a stronger Confucian culture will also have a stronger family culture. The weak IV test of Kleibergen–Paap F statistic presented at the bottom of Table 3 indicates the strength of instruments (Kleibergen and Paap, 2006).

Table 3.

IV estimates of family culture on marriage stability.

	(1)	(2)
	Family culture	Divorce
Number of memorial archways per 100,000 persons	0.2704***
	(0.025)
Family culture		−1.6103***
		(0.366)
Individual-level controls	Yes	Yes
Household-level controls	Yes	Yes
Community-level controls	Yes	Yes
Province-level controls	Yes	Yes
Intercept	−0.1911*	−2.8009***
	(0.105)	(0.768)
Kleibergen–Paap F statistics		121.13
Chi-squared	792.8142
Number of observations	23,041	23,041

Notes: standard errors in parentheses. * p < 0.1, *** p < 0.01.

Column (2) in Table 3 reports the second-stage results for the impact of family culture on marriage stability. The results show a significant negative correlation between family culture and marriage disruption, which means that family culture may play a protective role in the survival of a marriage.

Robustness checks

In this subsection, I will conduct a series of robustness checks for the impact of family culture on marriage stability.

Considering that the measurement of family culture by the visiting of graves at the family level may be biased, in order to more accurately measure the effect of family culture on marital stability, I will first conduct a robustness test using variables such as genealogy, the proportion of the most common surnames in the village, and the average frequency of visiting graves at the village level. First, in the existing literature, the proportion of the most common surnames is often used as a measure of clan culture. The basic logic is that the surname represents the basic clan relationship among the households, and the higher the proportion of the most common surnames, the closer the clan relationship and the stronger the clan culture. Chinese clan culture is also reflected in the importance that individuals and families attach to their genealogy. Second, families with a strong clan culture are more likely to preserve their genealogy and renew it regularly in order to pass on the family culture (C Zhang, 2017). Considering that the clan culture includes family culture and to a certain extent family culture is the continuation and transmission of clan culture in the nuclear family, the most important point is that these three variables all reflect ancestor worship, which is the core of family culture since they emphasize the importance of the patriarchy. Third, since individual behavior is susceptible to a variety of observable and unobservable factors such as the absence of ancestor worshiping ritual in the year of the interview, it thus may be unstable. Instead, the measurement bias may be smaller in a community average level because it is unlikely that all households in a community do not have ancestor worshiping ritual in the year of the interview. Therefore, this paper again measures the impact of family culture on marriage stability using family genealogy, the proportion of the most common surnames, and the average level of visiting graves at the village level.

The Probit results in Table 4 show that all three core explanatory variables are significantly negatively correlated with marriage disruption. These results are still consistent with the baseline results after controlling for the individual-level, household-level, and village-level variables and the results indicate that the probability of divorce is lower for the members in the family with genealogy, or those living in the village with a higher proportion of the most common surnames and higher degree of family culture at the community level.

Table 4.

Robustness checks—replacing the core explanatory variable.

	(1)	(2)	(3)	(4)	(5)	(6)
Genealogical tree	−0.205***	−0.108**
	(0.055)	(0.043)
Community-level family culture			−0.265***	−0.179**
			(0.075)	(0.089)
Proportion of the most common family names					−0.008***	−0.003**
					(0.001)	(0.001)
Individual-level controls	No	Yes	No	Yes	No	Yes
Household-level controls	No	Yes	No	Yes	No	Yes
Community-level controls	No	Yes	No	Yes	No	Yes
Intercept	−2.104***	−2.362***	−1.960***	−2.300***	−2.005***	−2.234***
	(0.023)	(0.312)	(0.055)	(0.311)	(0.024)	(0.313)
Pseudo R-squared	0.004	0.099	0.003	0.099	0.023	0.100
Chi-squared	14.767	1005.275	12.201	970.132	84.803	1419.844
Number of observations	23,041	23,041	23,041	23,041	22,851	22,851

Notes: standard errors in parentheses. Robust standard errors in columns (2), (4), (6) are clustered at the province level.

* p < 0.1, ** p < 0.05, *** p < 0.01.

Second, the paper examines the robustness of the benchmark results by replacing the model. The event-history analysis model not only estimates whether the event occurred or not, but also takes the duration before the event occurred into account, and this method can effectively deal with the problem of data censoring. Therefore, this paper applies event-history analysis (log-normal regression model) to estimate the effect of family culture on current marriage duration and first marriage duration. For those who are divorced now, the duration of marriage is calculated from the year of marriage with the former spouse to the year of divorce with the former spouse, and for those who are married now, the duration of marriage is observed from the year of marriage with the current spouse to the end of the observation period (the year 2010). Similarly, as for the first marriage, we begin observing respondents from the year of their first marriage. The observation stops at the year of divorce. If there is no occurrence of divorce, the observation ends at the interview (the year 2010). Column (1) and column (2) in Table 5 present the results of the effect of family culture on current marriage duration and first marriage duration, respectively. All the coefficients of the family culture in columns (1) and (2) are statistically significant, which means family culture significantly prolongs the duration of current marriage and first marriages and reduces the risk of divorce. It indicates that the baseline results are robust

Table 5.

Robustness checks—replacing the model and gender-stratified test.

	(1)	(2)	(3)	(4)
	Current marriage	First marriage	Male	Female
Family culture	0.309***	0.280***	−0.167***	−0.160**
	(0.105)	(0.098)	(0.062)	(0.072)
Individual-level controls	Yes	Yes	Yes	Yes
Household-level controls	Yes	Yes	Yes	Yes
Community-level controls	Yes	Yes	Yes	Yes
Intercept	2.668***	2.942***	−2.481***	−2.076***
	(0.884)	(0.799)	(0.489)	(0.573)
Chi-squared	551.9275	528.7596	189.8921	244.6596
Pseudo R-squared			0.0895	0.1476
Number of observations	21,894	21,258	11,310	11,731

Notes: standard errors in parentheses. ** p < 0.05, *** p < 0.01.

Third, considering that women are more likely to conceal their divorce due to the stigma of divorce in traditional Chinese society, this paper also conducts gender-stratified tests to verify the robustness of the basic results. The results are presented in columns (3) and (4) in Table 5. It shows that the effect of family culture on marital stability is significant in both male and female groups, also indicating that the basic result is robust.

Further research

The above analysis suggests that family culture has a positive impact on marriage stability. In this section, I will continue to investigate how this beneficial effect works in a different context. In particular, I examine the moderation role of perceived importance of family happiness, family size, and level of education.

Family well-being

The perception of family harmony is an important factor in the happiness and quality of marriage. From the theory of marital commitment (Johnson et al., 1999), an individual's family values are a personal commitment to marriage, and the degree to which an individual wishes to maintain an intact family and marital relationship. Family culture, on the other hand, is a structural commitment in which the spouses maintain the marital relationship because they are bound by objective external forces or social norms, such as culture, institutions, and the structure of the marriage market. Individual and structural commitments are not independent of each other but interact with each other. When an individual's marital commitment is high, he or she places more importance on the harmony and fulfillment of the marriage and family, and therefore his or her marriage lasts longer and has less risk of dissolution (Rusbult et al., 1998). Conversely, when individual commitment is low and they feel unhappy in the marriage, the structural commitment reflected in family culture comes into play.

Here, I will examine the moderation effect of the perceived importance of family well-being¹⁵ on marriage stability. Column (1) and column (2) of Table 6 show the results of interaction models that estimate whether the positive effects of family culture on marital relationships still exist in different degrees of family values. From column (1) we can observe that the coefficient of the interaction term is significantly negative, suggesting that the effect of family culture on marriage stability is moderated by the perceived importance of family well-being. Specifically, the effect of family culture on reducing divorce increases as the perceived importance of family well-being is weaker.

Table 6.

The moderating impact of perceived importance of family well-being.

	(1)	(2)
	Divorce	Divorce
Family culture	−0.049	−0.153***
	(0.081)	(0.056)
Family well-being	0.237***
	(0.051)
Family culture × family well-being	−0.077*
	(0.042)
Family well-being dummy		0.926***
		(0.151)
Family culture × family well-being dummy		−0.503***
		(0.164)
Individual-level controls	Yes	Yes
Household-level controls	Yes	Yes
Community-level controls	Yes	Yes
Intercept	−2.798***	−2.462***
	(0.277)	(0.297)
Pseudo R-squared	0.114	0.108
Chi-squared	1034.727	1210.001
Number of observations	23,041	23,041

Notes: Standard errors in parentheses. Robust standard errors are clustered at the province level.

* p < 0.1, *** p < 0.01.

To test the robustness of this interaction effect, I also constructed a dummy variable based on the perceived importance of family well-being (important = 0, not important = 1). The value of the perceived importance of family well-being harmony is assigned as 1 for values greater than 3, and 0 otherwise. The results in column (2) show an increase in the significance of the interaction term compared to column (1), further suggesting that the effect of family culture on marriage stability increases for individuals who perceive family well-being to be less important.

Family size

Family size is also an important moderating factor between family culture and marital stability. In China, marriage is not only a private affair between husband and wife but also a public affair involving the families of both parties. Therefore, marriage is affected by both individual commitments and family interventions. A couple's views about marriage and approach to intimacy will take into account the feelings and expectations of the other family members. The larger the family size, the greater the influence of the family and its power and authority over the marital relationship. For example, when a member of an extended family encounters setbacks and difficulties in his or her marriage, on the one hand, he or she will continue the marriage, taking into account the family's pride and the advice of family members, on the other hand, the family members will actively mediate between two parties to alleviate marital conflicts and tensions. On the contrary, for a small nuclear family, family forces interfere less in the marriage. When marital conflict arises in a small family, the binding effect of the family culture of harmony on marriage also begins to emerge. In other words, the smaller the family size, the stronger the effect of the family culture on the stability of the marriage.

Next, I will investigate the role of family size. The results of the interaction term presented in Table 7 indicate that the positive effect of family culture on marriage stability will decrease with the enlarging of family size. This result is also existing after controlling all the other individual-level, household-level, and village-level variables.

Table 7.

The moderating impact of family size.

	(1) Divorce	(2) Divorce
Family culture	−0.409***	−0.481**
	(0.126)	(0.216)
Family size	−0.340***	−0.355***
	(0.032)	(0.052)
Family culture × family size	0.086**	0.112*
	(0.039)	(0.065)
Individual-level controls	No	Yes
Household-level controls	No	Yes
Community-level controls	No	Yes
Intercept	−0.877***	−0.293
	(0.102)	(0.278)
Pseudo R-squared	0.094	0.159
Chi-squared	355.779	1529.140
Number of observations	23,041	23,041

Notes: standard errors in parentheses. Robust standard errors in column (2) are clustered at the province level.

* p < 0.1, ** p < 0.05, *** p < 0.01.

Education

The existing literature found that education has a strong inhibitory effect on traditional culture. The improvement of educational attainment enables individuals to be exposed to more open, pluralistic, egalitarian, and liberal values. Therefore, I expect education to play a moderating role between family culture and marital stability. That is, the influence of family culture on marriage stability will increase for people with a lower education level.

Table 8 presents the results of the interaction of education¹⁶ and family culture. The interaction term is significantly negative in column (1), which indicates that the positive effect of family culture on marital stability increases with educational attainment decreasing. To test the robustness of this result, I also define a sample with an education level of junior school and below as a low-educated sample, assigning a value of 1 to this sample, and 0 otherwise, to construct a dummy variable of education degree (low level of education = 1, high level of education = 0). The results in column (2) show a stronger significance of interaction term than in column (1).

Table 8.

The moderating impact of education.

	(1) Divorce	(2) Divorce
Family culture	0.236	−0.012
	(0.221)	(0.078)
Education	0.027
	(0.038)
Family culture × education	−0.066**
	(0.032)
Education dummy		0.106
		(0.076)
Family culture × education dummy		−0.218***
		(0.082)
Individual-level controls	Yes	Yes
Household-level controls	Yes	Yes
Community-level controls	Yes	Yes
Intercept	−2.532***	−2.448***
	(0.464)	(0.305)
Pseudo R-squared	0.103	0.103
Chi-squared	1170.875	1684.571
Number of observations	23,041	23,041

Notes: standard errors in parentheses. Robust standard errors are clustered at the province level. ** p < 0.05, *** p < 0.01.

Birth cohort distinction

So far, we have seen how the positive effect of family culture on marriage stability changes in different family environments and in line with individual education. Finally, I will examine whether this positive effect of family culture on marriage stability will be existing in a different birth cohort. The results are presented in Table 9, indicating that this positive role of family culture on marriage stability is mainly only seen in those people who were born before 1969. Since the older generation had grown up in a more traditional and closed cultural environment, while the younger generation is affected by modernization and accepts more value of equality and freedom, therefore their marriage values and relationships are more open.

Table 9.

Birth cohort difference.

	(1)	(2)	(3)	(4)
	Before 1959	1960–1969	1970–1979	1980–1990
Family culture	−0.156**	−0.276***	−0.129	0.022
	(0.077)	(0.089)	(0.093)	(0.182)
Individual-level controls	Yes	Yes	Yes	Yes
Household-level controls	Yes	Yes	Yes	Yes
Community-level controls	Yes	Yes	Yes	Yes
Intercept	3.187	15.611	5.768	−3.115
	(2.227)	(12.164)	(7.825)	(10.554)
Pseudo R-squared	0.109	0.137	0.127	0.146
Chi-squared	157.651	144.391	125.568	43.185
Number of observations	9332	6140	5153	2416

Notes: standard errors in parentheses. Robust standard errors are clustered at the province level. * p < 0.1, ** p < 0.05, *** p < 0.01.

Conclusion and discussion

This paper examines the effect of family culture on marriage stability. Using the CFPS data, I first show the large variation in family culture across the different regions in China. Following this, I found that family culture has a positive impact on marriage stability. To mitigate the omitted variables and measurement error problems, I adopted the IV approach. Using a unique dataset on the distribution of memorial archways in China, I instrumented the number of the memorial archways per 100,000 persons in each province and the results of the IVs support the main conclusion. Finally, the robustness checks are all still consistent with the major finding.

I then test the moderating role of the perceived importance of family well-being, family size, and education. The results show that the positive impact of family culture on marriage stability is mainly seen in those people with a weaker conception of family well-being and lower level of education and for individuals living in a small family.

Finally, as modernization and industrialization gradually deconstruct traditional culture, the positive effect of family culture on marital stability is diminishing for younger generations. Specifically, this positive effect is only observed in older generations, especially for the people born before 1969.

Some factors can to some extent explain the rising divorce rate in the younger generation recently in China. First, the increase of women's educational attainment and labor market participation enhanced their economic power and independence. Therefore, the traditional division of gender roles in the family has been weakened. This gives women more confidence to end a marriage. Also, the “de-stigmatization” of divorce and the rising social tolerance for divorce and remarriage makes individuals in a marriage less family-centered and more individual-centered. Thus, when a marriage breaks down, the man and woman may divorce in pursuit of their happiness and freedom. Third, with the opening of marriage concepts and the diversification of spouse selection, social norms have become less binding on extramarital sex.

The family is the cell of society. Since the second demographic transition, the size and structure of the family are increasingly nuclear, and the family has become loose and fragile under the influence of modernization and individualism. The functions of the family in terms of marriage, emotions, childbirth, and old-age support are getting weaker and blurred, and some scholars call it the “mosaic family” (Ji, 2015, 2017). Marital instability exacerbates the weakening of family functions and has a negative impact on parental well-being (Osborne et al., 2012), the growth of children (Amato, 2000), and social stability. On the contrary, family integrity plays an important role in the development of children, elder support, and family prosperity. The stability of marriage is the foundation and key to family construction and social development. This study demonstrates the positive role of Chinese traditional family culture, which promotes family harmony, in maintaining marital stability and it also shows the coexistence of tradition and modernity in the Chinese family. Although the foundation of family culture was dissolved with modernization, the social normative role still played by a family culture still regulates Chinese families (Li et al., 2020). The value of harmony and subordination advocated by family culture and the obstacle of family forces to divorce can still have a powerful influence on the stability of modern marriages and show its strong vitality. Therefore, to re-establish the function of the family, it is even more important to repair marriages and to pass on and promote the best values of the family culture.

As an exploratory study of the influence of culture on family and marital behavior, this paper has some limitations. First, due to data limitations, I use ancestor worship, which was measured as visiting graves regularly, as a proxy variable for family culture, and as such this is not a direct measure of family culture. Since the conception of culture is abstract and multidimensional, there may be some measurement bias. However, this study also makes an effort to try to mitigate endogeneity issues brought by measurement error and omitted variables using a variety of methods such as replacement variables and IV methods, and the robustness results all support the basic results. In addition, this study also attempts to indirectly explore the mechanisms by which family culture affects marital stability and finds the moderation role of values of family well-being and family size, but due to data limitations, a test of the intermediation mechanisms is lacking. I also look forward to further empirical testing of intermediary mechanisms in the future with more comprehensive data.

Footnotes

Acknowledgments

The author thanks Prof. Yuan Zhao, Qin Ma, and Prof. Lisha Hao for providing the dataset of memorial archways. The author also thanks Dr. Huan Zhang for his helpful comments and suggestions.

Declaration of conflicting interests

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

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Social Sciences Foundation (Grant numbers: 17ARK002 and Shanghai Municipal Education Commission Research and Innovation Project (Grant numbers: E00026).

ORCID iD

Li Hao

Notes

References

Amato

(2000) The consequences of divorce for adults and children. Journal of Marriage and Family 62(4): 1269–1287.

Amato

(2010) Research on divorce: Continuing trends and new developments. Journal of Marriage and Family 72(3): 650–666.

Amato

Irving

(2006) Historical trends in divorce in the United States. In: Fine

Harvey

(eds) Handbook of Divorce and Relationship Dissolution. Hove: Psychology Press, 41–57.

Axinn

Thornton

(1992) The relationship between cohabitation and divorce: Selectivity or causal influence? Demography 29(3): 357–374.

Bramlett

Mosher

(2002) Cohabitation, Marriage, Divorce, and Remarriage in the United States. Vital and Health Statistics series, Series 23, Number 22, July. Hyattsville, MD: Centers for Disease Control and Prevention.

Bumpass

(1990) What's happening to the family? Interactions between demographic and institutional change. Demography 27(4): 483–498.

Chen

(2018) Flowing social capital: Can traditional clan culture affect migrants’ employment in modern society? Economic Research Journal 53(3): 35–49 (in Chinese).

Chen

(2019) Social norms and household savings rates in China. Review of Finance 23(5): 961–991.

Clark

(1999) Mothering, work, and gender in urban Asante ideology and practice. American Anthropologist 101(4): 717–729.

10.

Cohen

(1990) Lineage organization in north China. Journal of Asian Studies 49(3): 509–534.

11.

De Paola

Gioia

(2017) Does patience matter in marriage stability? Some evidence from Italy. Review of Economics of the Household 15(2): 549–577.

12.

Dzara

(2010) Assessing the effect of marital sexuality on marital disruption. Social Science Research 39(5): 715–724.

13.

Eeckhaut

Lievens

Putte

BVD

, et al. (2011) Partner selection and divorce in ethnic minorities: Distinguishing between two types of ethnic homogamous marriages. International Migration Review 45(2): 269–296.

14.

Fei

Chang

Cooper

, et al. (1975) Earthbound China. Chicago, IL: University of Chicago Press.

15.

Fernández

(2011) Does culture matter? In: Benhabib

Bisin

Jackson

(eds) Handbook of Social Economics. Amsterdam: Elsevier, pp. 481–510.

16.

Fernández

Fogli

(2009) Culture: An empirical investigation of beliefs, work, and fertility. American Economic Journal: Macroeconomics 1(1): 146–177.

17.

Freedman

(1966) Chinese Lineage and Society: Fukien and Kwangtung. London: Athlone Press.

18.

Fukuyama

(1999) The Great Disruption: Human Nature and Reconstitution of Social Order. London: Profile Books.

19.

Furtado

Marcén

Sevilla

(2013) Does culture affect divorce? Evidence from European immigrants in the United States. Demography 50(3): 1013–1038.

20.

Greif

Tabellini

(2017) The clan and the corporation: Sustaining cooperation in China and Europe. Journal of Comparative Economics 45(1): 1–35.

21.

Guiso

Sapienza

Zingales

(2006) Does culture affect economic outcomes? Journal of Economic Perspectives 20(2): 23–48.

22.

Heaton

(2002) Factors contributing to increasing marital stability in the United States. Journal of Family Issues 23(3): 392–409.

23.

Hitlin

Piliavin

(2004) Values: Reviving a dormant concept. Annual Review of Sociology 30: 359–393.

24.

Hou

(1998) Population survey in the Xuantong period: A review of Mi Hong et al.'s Paper and other related studies. Historical Research 6: 134–147 (in Chinese).

25.

Hsu

(1971) Under the Ancestors’ Shadow: Kinship, Personality, and Social Mobility in Village China. Stanford, CA: Stanford University Press.

26.

Hsu

(1981) Americans and Chinese: Passages to Differences. Honolulu: University of Hawaii Press.

27.

(2020) Are elders from ancestor-worshipping families better supported? An exploratory study of post-reform China. Population Research and Policy Review 40(3): 1–24.

28.

Huang

Yuan

(2014) A study on equality and GIS spatial distribution of educational attainment growth of China: 1982–2010. Population Journal 36(5): 5–17 (in Chinese).

29.

Jensen

Smith

(1990) Unemployment and marital dissolution. Journal of Population Economics 3(3): 215–229.

30.

(2015) Between tradition and modernity: “Leftover” women in shanghai. Journal of Marriage and Family 77(5): 1057–1073.

31.

(2017) A mosaic temporality: New dynamics of the gender and marriage system in contemporary urban China. Temporalités 26(1): 1–16.

32.

Jin

(2003) Origin, development, and social functions of memorial archways. Forum on Chinese Culture 1: 71–75 (in Chinese).

33.

Johnson

Caughlin

Huston

(1999) The tripartite nature of marital commitment: Personal, moral, and structural reasons to stay married. Journal of Marriage and the Family 61(1): 160–177.

34.

Kleibergen

Paap

(2006) Generalized reduced rank tests using the singular value decomposition. Journal of Econometrics 133(1): 97–126.

35.

Kung

(2014) Can cultural norms reduce conflicts? Confucianism and peasant rebellions in Qing China. Journal of Development Economics 132–149.

36.

Fan

Song

(2020) The household structure transition in China: 1982–2015. Demography 57(4): 1369–1391.

37.

(2018) Migration and marital instability among migrant workers in China: A gender perspective. Chinese Journal of Sociology 4(2): 218–235.

38.

Oppong

(1974) Marriage Among a Matrilineal Elite: A Family Study of Ghanaian Senior Civil Servants. London: Cambridge University Press.

39.

Osborne

Berger

Magnuson

(2012) Family structure transitions and changes in maternal resources and well-being. Demography 49(1): 23–47.

40.

Parkman

(2000) Good Intentions Gone Awry: No-Fault Divorce and the American Family. Lanham, MD: Rowman and Littlefield.

41.

Parsons

Shils

(1951) Toward a General Theory of Action. Cambridge, MA: Harvard University Press.

42.

Peng

(2004) Kinship networks and entrepreneurs in China's Transitional economy. American Journal of Sociology 109(5): 1045–1074.

43.

Rogers

(2004) Dollars, dependency, and divorce: Four perspectives on the role of wives’ income. Journal of Marriage and Family 66(1): 59–74.

44.

Rusbult

Martz J

Agnew

(1998) The investment model scale: Measuring commitment level, satisfaction level, quality of alternatives, and investment size. Personal Relationships 5(4): 357–387.

45.

Santos

Harrell

(2017) Transforming Patriarchy: Chinese Families in the Twenty-First Century. Seattle, WA: University of Washington Press.

46.

Takyi

Gyimah

(2007) Matrilineal family ties and marital dissolution in Ghana. Journal of Family Issues 28(5): 682–705.

47.

Thornton

Young-DeMarco

(2001) Four decades of trends in attitudes toward family issues in the United States: The 1960s through the 1990s. Journal of Marriage and Family 63(4): 1009–1037.

48.

Tsai

(2007) Accountability Without Democracy: Solidary Groups and Public Goods Provision in Rural China. Cambridge: Cambridge University Press.

49.

Wang

(2008) The theoretical analysis for the family structure transition: Based on the historical and empirical evidence in rural China. Journal of Social Science 7: 90–103 (in Chinese).

50.

Weber

(1951) The Religion of China: Confucianism and Taoism (Trans. by Gerth HH). New York: Macmillan.

51.

Wolfers

(2006) Did unilateral divorce laws raise divorce rates? A reconciliation and new results. American Economic Review 96(5): 1802–1820.

52.

Wolfinger

(2000) Beyond the intergenerational transmission of divorce: Do people replicate the patterns of marital instability they grew up with? Journal of Family Issues 21(8): 1061–1086.

53.

Zhao

Xie

(2020) Marriage and divorce in China: Trends and global comparison. Population Research 44(5): 3–18 (in Chinese).

54.

Zhang

(2017) Culture and the economy: Clan, entrepreneurship, and development of the private sector in China. SSRN Electronic Journal. Available at: https://ssrn.com/abstract = 2865105 or https://dx-doi-org.web.bisu.edu.cn/10.2139/ssrn.2865105 (accessed 9 September 2019).

55.

Zhang

(2019) Family support or social support? The role of clan culture. Journal of Population Economics 32(2): 529–549.

56.

Zhang

(2011) Story of Chinese Memorial Archways. Jinan: Shandong Pictorial Publishing House.

57.

Zhang

(2017) Premarital cohabitation and marital dissolution in postreform China. Journal of Marriage and Family 79(5): 1435–1449.

58.

Zhao

Hao

(2016) Regional differentiation of existing Chinese memorial archways and its causes. Geographical Research 35(10): 1948–1962 (in Chinese).