Search tips
Search criteria 


Logo of geronbLink to Publisher's site
J Gerontol B Psychol Sci Soc Sci. 2013 May; 68(3): 420–430.
Published online 2013 April 11. doi:  10.1093/geronb/gbt022
PMCID: PMC3627660

Sources of Older Parents’ Ambivalent Feelings Toward Their Adult Children: The Case of Rural China



Relying on the concept of sociological ambivalence, this study investigated the sources of intergenerational ambivalence in rural China, a traditional society that undergoes dramatic demographic, economic, and social changes.


Data were derived from a survey of 1,162 older adults and their 4,396 children in rural Anhui province. Two-level mixed-effects model was carried out to predict ambivalent feelings reported by the older parents toward their adult children.


The findings yielded both similar and distinctive predictors of intergenerational ambivalence among the Chinese elderly population compared with their Western counterparts. The Chinese elderly population reported greater ambivalence toward sons than toward daughters. Adult children’s higher socioeconomic status, represented by a more prestigious job and an urban household registry status (hukou), was associated with reduced ambivalence among parents. Parents’ monetary support to children and assistance with childcare, which is common in rural China due to the massive out-migration, were also associated with higher levels of ambivalent feelings.


We interpreted the findings in the larger social context of strong son preference, large-scale rural-to-urban migration, and rigid rural–urban division in China due to the household registration system. Our findings demonstrate that individual feelings of ambivalence are culturally structured and are determined within complex social environment.

Key Words: China, Family relations, Intergenerational ambivalence, Local context—, Sociological ambivalence.

Although family relationships are often characterized by positive affect, such as affection, mutual support, and family obligations, it is not uncommon that family members also experience conflicting relationships such as tensions, disappointments, and unpleasantness (Bengtson, Rosenthal, & Burton, 1996; Birditt, Jackey, & Antonucci, 2009; Krause & Rook, 2003). A considerable proportion of parents and adult children also experience mixed feelings (Fingerman, 1998; Pillemer & Suitor, 2002). Researchers agree that the feelings of ambivalence, the paradox between closeness and distance, and the paradox between intimacy and setting boundaries are inevitable in intimate social relations such as those in families (Bengtson, Giarrusso, Mabry, & Silverstein, 2002; Fingerman & Hay, 2004). The concept of intergenerational ambivalence, defined by Lüscher (2002) as “the simultaneous coexistence and opposition of harmony and conflict in intergenerational relations” (p. 591), overcomes the “either solidarity or conflict” simplicity of family relationships, depicting a more completed picture of family dynamics.

To date, our current knowledge about intergenerational ambivalence primarily comes from the findings of Western studies (Fingerman, Chen, Hay, Cichy, & Lefkowitz, 2006; Ha & Ingersoll-Dayton, 2008; Willson, Shuey, & Elder, 2003; Willson, Shuey, Elder, & Wickrama, 2006). It is rarely studied whether the Western-derived concept of ambivalence can be applied to traditional families like those in Asia. To fill this research gap, this study investigates the sources of intergenerational ambivalence among older adults in rural China, a traditional society that undergoes dramatic demographic, economic, and social changes. We are interested in examining whether the different socioeconomic and normative contexts of Western and Eastern families produce different predictors of intergenerational ambivalence.

Dimensions of Ambivalence

According to Lüscher and Pillemer (Lüscher, 2002; Lüscher & Pillemer, 1998), ambivalence is experienced at two levels, or it has two dimensions. Psychological ambivalence is defined as individuals’ conflicting cognitions, emotions, or attitudes resulted from simultaneously opposing feelings toward the same object (Lowenstein, 2007; Lüscher & Pillemer, 1998). It emphasizes the feeling states and emotional experience at the individual level and has been frequently used in studies of clinical psychology (see Smelser, 1998 for a review). In contrast, sociological ambivalence is conceptualized at the society level and is produced as a result of “incompatible normative expectations of attitudes, beliefs, and behavior” (Merton & Barber, 1963, p. 95). The premise of sociological ambivalence is that when individuals possess roles or positions that require contradictory expectations for behaviors, the pressure to subscribe to these incompatible norms produces ambivalence (Connidis & McMullin, 2002b; Fingerman, Pitzer, Lefkowitz, Birditt, & Mroczek, 2008). A classic example of sociological ambivalence is conflicting expectations of physicians to be detached and professional yet compassionate (Grzywacz et al., 2006). In family studies, gender and generation are important structural sources of ambivalence. Women are more likely than men to experience ambivalence due to conflicting normative expectations of caregiving in the domestic domain and increasing expectations of employment in the economic domain (Connidis & McMullin, 2002a; Fingerman et al., 2008). Parents are likely to experience ambivalence when the norms of intergenerational solidarity (e.g., helping adult children) and intergenerational independence (e.g., being freed from their demands) collide (Levitzki, 2009; Pillemer & Suitor, 2002).

However, social structures produce ambivalence by more than creating norms that require contradictory attitudes and behaviors. A central, yet relatively understudied aspect of the sociological conceptions of ambivalence is how the broader economic, social, and cultural structures form a constraining world within which individuals as social actors attempt to exert control over their lives. Individuals may experience ambivalence when structural arrangements constrain their ability to fulfill expectations or to exercise agency in negotiating relationships with others, including family members (Connidis & McMullin, 2002b). Examples of such sociological ambivalence are ample in international and cross-cultural studies that are situated in changing social contexts. For instance, studies of both internal and international migrations show that migrants, particularly married ones, usually have contradictory feelings about the movement because it often compromises family relationships and families’ well-being, yet the movement is needed because adequate employment is not available in the local community (Grzywacz et al., 2006; Guo, Chow, & Palinkas, 2011). Ambivalent feelings are also observed among older adults who experience changing life circumstances and cultural dissonance in family relations. Older immigrants in the United States often experience ambivalence because their traditional beliefs and behaviors of intergenerational relations are challenged, stressed, and altered in the new society, being no longer sustainable (Lewis, 2008; Treas & Mazumdar, 2002). Older Japanese in Japan face similar dilemmas as their expectations of family relations cannot be fulfilled in rapid modernization (Traphagan, 2010). These studies exemplify how the local context constrains or shapes individual’s ability in negotiating and managing family relations, which in turn result in ambivalent feelings. These findings extend sociological conception of ambivalence that tends to be restricted to competing normative systems, providing a new avenue for discerning the associations between social structures and ambivalent feelings.

In sum, although closely related, the two dimensions of ambivalence differ in their applications in empirical studies. Whereas psychological ambivalence is suitable to study the measurement of ambivalence, sociological ambivalence can be a bridging concept between social structure and individual action and is tuned to studies that link individual sentiments to their roots in structural arrangements (Willson et al., 2003). Sociological ambivalence is particularly relevant to this study. If the development of ambivalent feelings is conditioned on the larger social and normative context, one may expect to find distincive predictors of ambivalence among Chinese rural elders in comparision with older adults in the West.

Sources of Intergenerational Ambivalence: Evidence From the West

Empirical studies of intergenerational ambivalence in the West mainly focus at the individual level and emphasize subjective emotions and feelings from the psychological perspective. Empirical findings identify three aspects of parent–child relations that are likely to generate ambivalent feelings among family members: (a) gender-patterned family relations, (b) children’s life circumstances, and (c) intergenerational dependence.

Mother–daughter dyads usually rank higher on levels of ambivalence compared with dyads including men (Willson et al., 2003). This is because women usually invest more time and emotions in their families than do men, and thus are more likely to experience both affectionate and distressful emotional intensity in family relations, being more likely to develop intergenerational ambivalence (Fingerman, 2001).

Adult children’s life circumstances that are viewed as “unsuccessful” or “unfavorable” by the parents tend to trigger parents’ ambivalent feelings. To assess children’s successes, researchers typically examine children’s socioeconomic status (SES; e.g., education attainment, career success, and financial status) and social roles (e.g., being married, having children). Parents report greater ambivalence toward children who have lower SES (not completing college, having financial difficulties) or who have not achieved adult status (married, having children; Birditt, Fingerman, & Zarit, 2010; Fingerman et al., 2006; Pillemer & Suitor, 2002). There is also higher level of ambivalence among parents whose children have unwise choice of romantic partners or unfavorable parenting styles (Peters, Hooker, & Zvonkovic, 2006). It is explained by the earlier studies that when children fail to meet these normative expectations, ambivalent feelings develop among parents as a result of mixed feelings of affections to the child and disappointment at the child’s situation or even self-doubt about parenting.

Intergenerational dependence is likely to generate ambivalent feelings through contradictory norms of solidarity and reciprocity. Mixed feelings are evident among older parents whose adult children remain dependent upon them for financial support (Ingersoll-Dayton et al., 2011). Older adults who are custodial grandparents also report being caught between the desire to help their children and grandchildren and their reluctance to take on the parental role again and give up plans for retirement (Letiecq, Bailey, Dahlen, Hayslip, & Kaminski, 2008). Overall, the earlier studies shed important light on ambivalence as a manifestation of contrasting feelings and expectations, but they center predominantly on attributes of individuals and their interpersonal relationships at the expense of broader sociostructural contexts (Hillcoat-Nallétamby & Phillips, 2011). Individuals are social actors who cannot be reduced only to psychological states and feelings (Connidis & McMullin, 2002a). It is the local context of this study that we now turn our discussion to.

Rural China: The Local Context

Traditional Chinese families, particularly those in rural areas, have long been characterized as patriarchal (Zhan & Montgomery, 2003). In rural China, sons bear the major responsibilities of old-age care, both legally and culturally, in forms of coresidence, food provision, and instrumental and financial support (Xu, 2001). Compared with sons, daughters are regarded to have much less to offer. They ultimately join the husband’s lineage upon marriage, after which, their obligations are directed toward in-laws rather than their own parents (Whyte & Qin, 2003). Xu (2001) wittily summarized that in rural China “support by sons is like pensions that the aged parents are entitled to . . . , whereas support by daughters is a bonus, which is . . . an additional blessing” (p. 131). The strong son preference in rural China has significant implications for potential sources of intergenerational ambivalence. Chinese parents, particularly those in rural areas, tend to have higher expectations on sons than on daughters, both in terms of old-age support in the family domain and individual’s success in the economic domains. The higher expectations on sons are likely to result in greater ambivalence among parents toward sons when these expectations are not fulfilled. In contrast to the daughter-centered family relations in the West, one would also expect more frequent interactions between parents and sons than between parents and daughters, which may foster affection and conflict, consequently generating greater ambivalence among dyads with sons than dyads with daughters in China.

Rural-to-urban migration is another formidable structural context that shapes family dynamics in rural China. Chinese rural-to-urban migration soared more than sixfold from 34 million in 1989 to 224 million in 2010, making up roughly one third of the total Chinese rural population (National Bureau of Statistics of China, 2011). The influence of such unprecedented migration on rural families is profound. One notable change is in living arrangement. Due to the high cost of living in cities, most rural migrants move on their own, leaving their children in the home villages to be cared by their older parents. There are increasing skip-generation households in rural China, in which older elders take the primary childcare responsibilities in lieu of their absent children (Jingzhong & Lu, 2011). Such a change alters the traditional intergenerational patterns in China because rural elders, who usually have limited resources and rely on adult children for various support, become a major source of instrumental assistance to their children. Although the grandparenting role may create a sense of fulfillment among rural elderly population, it is often long term, demanding, and may leave older adults with few options of exit, leading to a sense of ambivalence (Guo, Chi, & Silverstein, 2012).

A less direct but similarly profound influence of the rural-to-urban migration on rural families is involving parental expectations on adult children in terms of success and desired life circumstances. Rural-to-urban migration provides rural residents, particularly young people, access to better employment and higher income that are not available in the local community. The seemingly promising opportunity, however, is largely restrained by the household registration (hukou) system in China, which designates each Chinese resident with a record of registration location of either urban or rural areas (Chen & Liu, 2009). The importance of hukou system goes beyond types of residence because it largely determines one’s eligibility and access to various social and economic benefits that are largely tied up with the urban hukou. Without an official urban hukou, rural residents working and living in cities are unlikely to find a stable job, to send their children to local schools, or to claim many social benefits such as housing, pensions, and health care (Fan, 2002). Given the privileges associated with the urban hukou, we argue that having an urban hukou may be an involving indicator of economic and social achievement among rural residents that is not captured by the common measures of SES such as income and wealth. Because intergenerational ambivalence develops when children are less successful or do not achieve desired social and economic status, we expect that children’s urban hukou status is associated with a reduced level of ambivalence among older parents.

Overall, the unique sociocultural contexts of rural China lead us to develop three hypotheses about the potential sources of intergenerational ambivalence in rural China. First, given the strong son preference in rural China and consequently greater expectations on sons and greater interactions between parents and sons, we hypothesized that

H1: Older Chinese parents experience a greater level of ambivalence toward sons than toward daughters.

The second hypothesis relates to the hukou status, a unique indicator of SES in China. We hypothesized that

H2: Chinese older parents experience a lower level of ambivalence toward children who have better SES, particularly, who have an urban hukou status.

Finally, given that ambivalence increases under conditions of potential dependence and given the increasing number of skip-generation households in rural China, we hypothesized that

H3: Providing monetary support and childcare assistance to their children are associated with greater ambivalence among Chinese older parents.



The data used for this study were derived from Anhui province, an underdeveloped province with massive out-migration. In 2001, multistage cluster sampling was used to select 1,802 adults aged 60 and older from the administrative records of 72 randomly selected villages within six rural townships in Chaohu region, Anhui. Within each selected village, potential respondents were stratified to form two sampling frames based on age: (a) those aged 60–74 and (b) those aged 75 and older. Two measures were adopted in the sampling procedure to guarantee that only one older person per household was interviewed. If household partners were from the two age groups, the older partner was interviewed. Another older adult aged between 60 and 74 was selected randomly from the village as a replacement. If household partners were from the same age group, then the partner chosen first was interviewed. Surveys were conducted in respondents’ homes. Respondents discussed each child at length in various family relations. Follow-up surveys were administered to the original respondents in 2003, 2006, and 2009. The 2009 survey also recruited 416 new participants aged 60 and older from the same villages. The data used in this study were derived from the 2009 survey because a key study variable—conflictual family relations—was only available in this survey. The 2009 survey consisted of 1,179 older adults who reported information of 4,472 children. We excluded the cases of deceased children (n = 23) and children who had missing information on parent–child relationships (n = 53). This led to a working sample of 1,162 older adults with 4,396 children.


Intergenerational ambivalence.—

Based on the concept of psychological ambivalence (i.e., simultaneous positive and negative feelings toward the same object), intergenerational ambivalence was assessed by asking respondents separate questions about positive and negative components of their relationships with each child. The positive dimension was assessed by three questions: “how close do you feel toward this child?”, “how much do you feel that this child would be willing to listen when you need to talk about your worries and problems?”, and “overall, how well do you and this child get along together?” (0 = not at all, 1 = somewhat, and 2 = very). Similarly, the negative dimension was assessed by three questions: “how often do you have tense and strained feelings with this child?”, “how often do you think this child makes too many demands on you for help and support?”, and “how often do you feel that this child is critical of you or what you do?” (0 = never, 1 = rarely or sometimes, and 2 = often). Two composite scores, one measuring cohesion and one measuring conflict, were calculated for each child, both ranging from 0 to 6, with higher scores representing closer/more conflictual parent–child relations. Reliability of the cohesion and the conflict scales were α = 0.95 and α = 0.67, respectively. We calculated the ambivalence score using Thompson formula (Thornton, Fricke, & Yang, 1994), which has been widely used in studies of intergenerational ambivalence (Birditt et al., 2010; Fingerman et al., 2008; Ha & Ingersoll-Dayton, 2008; Willson et al., 2006).

An external file that holds a picture, illustration, etc.
Object name is geronb_gbt022_m0001.jpg

The first part of the formula shows that both the feelings of positive and negative should be intense in order to obtain a high ambivalence score. Otherwise, there is absence of feelings or the relationships are indifferent. The second part of the formula suggests that the presence of ambivalence is evident when positive and negative dimensions are similar in magnitude. If the magnitude differs, the feelings become increasingly polarized in the direction of the stronger component. The ambivalence score was calculated for each parent–child dyad, ranging from −3 to 6. A higher score represents greater ambivalence toward the child.

Children’s gender.—

Children’s gender was coded as a dummy variable, with son = 1.

Children’s SES.—

In addition to the common measures of SES such as education, occupation, and financial conditions, we added household registration (hukou) status as a unique indicator of children’s SES in this study. Based on education of children, we had four categories: no formal education (reference group), completed primary school, completed junior high school, and completed senior school and above. Similarly, based on their occupation, we had four categories: farmers (reference group), workers, professionals or entrepreneurs, and others. Economic status was assessed by parents’ report of each child’s financial conditions: 1 = having some economic difficulties and 0 = no economic difficulties. Hukou status was coded as a dummy variable, with 1 = urban hukou status. The correlations among the four SES indicators ranged from 0.02 to 0.40, suggesting that each measure represented a distinctive dimension of the SES.

Parents’ support to their children included monetary support and help with childcare in the past year. To assess children’s “dependency” on the parents, monetary support was indicated by the net monetary transfer from a parent to the child in the prior year, calculated by subtracting the amount of money received from a child from the amount of money given to the child. The value ranged from −25,000 to 54,960 RMB. Because the distribution of this variable was skewed, we used the natural log of the RMB value (+25,001) to represent this variable. Help with childcare was a dummy variable, with 1 = provided childcare in the past year.

Control variables were characteristics of parents and children that have been shown to affect intergenerational relations in previous studies. Parents’ characteristics included age, gender, marital status, education, income, and functional health status. Age was measured in years. Gender and marital status were coded as dummy variables, with female = 1 and married = 1. Due to the low education attainment of Chinese rural elderly population, education of the parents was coded as a dummy variable, with received any formal education = 1. Income of parents was indicated by the total income that the respondent (and spouse, if married) had received from work or pension during the past year. The natural log of the Chinese RMB currency value (+1) was used to represent income level. Health status of the parents was represented by respondents’ level of difficulty (0 = no difficulty, 1 = some difficulty, 2 = cannot perform without help) in performing 15 tasks that represented activities of daily living (dressing/undressing, bathing, walking around the room, getting out of bed/chair, going to the toilet, eating); instrumental activities of daily living (preparing meals, shopping, doing housework, taking the bus or train, managing money); and activities requiring physical strength, mobility, and flexibility (lifting a 10 kg bag of rice, climbing one flight of stairs, walking 100 m, stooping/crouching/kneeling).The summed functional health difficulty scores ranged from 0 (no difficulties performing any task) to 30 (unable to perform any task), with a reliability coefficient of 0.93.

Children’s characteristics included age (in years), marital status (1 = married), geographic distance from the parent (ranging from 1 = living together to 7 = living in different provinces), and the frequency of face-to-face contact with the parent (ranging from 1 = seldom to 7 = daily).

Data Analysis

Descriptive data were presented to provide a profile of the parents and the children in this study. Because the dependent variable—ambivalent feelings in each parent–child dyad—was nested within parents, two-level regression models were estimated to study the influence of both children’s (Level 1) and parents’ (Level 2) characteristics on ambivalence. The two-level regression models were specified as random intercept models, allowing parent-specific intercepts to vary randomly from family to family. By doing so, we assumed that in addition to parents’ characteristics included in the models, there are additional unmeasured effects at the family level that explain between-family variation in intergenerational ambivalence. For ease of interpretation, continuous predictors at Level 1 (i.e., children’s age and net monetary transfer from parents to children) were centered on the mean of all children within the same families.

The multilevel analysis was run in three steps (de Leon & Glass, 2004). The first step involved running an empty model to partition the total variability in intergenerational ambivalence into its within- and between-family components. The second step examined the effect of Level 1 variables (i.e., children’s attributes and parents’ support to each child) on ambivalence scores. The third step determined the effect of Level 2 variables (i.e., parent’s attributes) on ambivalence scores. Using the approach advocated by Albright (2007), the analysis was carried out in STATA using.xtmixed command (StataCorp, 2005).


Table 1 presents the characteristics of the sample. The mean age of the parents and the children were 72 and 42 years, respectively. There were slightly more mothers (52%) than fathers and more sons (53%) than daughters in the sample. The majority of the parents (64%) and the children (79%) were married. The parents were more than twice as likely as the children to be illiterate (67% vs. 27%). The average annual income of the parents was 2,443 RMB (equals to 343 USD). The parents reported relatively low functional health difficulties (M = 3.88 out of a possible 30, SD = 6.77). Farmers (42%) and workers (45%) were the most common occupations of the children. About 9% of the children were professionals or entrepreneurs and 10% of the children had urban hukou status. According to the parents, 15% of the children had some financial difficulties. On average, children provided more monetary support to their parents than they received from parents. About 19% of the parents helped their children with childcare.

Table 1.
Sample Characteristics of the Older Parents (N = 1,162) and Their Children (N = 4,396) in Anhui Study (2009)

Table 2 shows the results of the multilevel analyses. The empty model shows that the average ambivalence scores were −1.09 across families. The variability of the family mean was 2.53 and the residual was 1.53, resulting in an intraclass correlation of 0.62. This suggests that the majority (62%) of the variation in parents’ ambivalent feelings was between families.

Table 2.
Multilevel Regression Results for Intergenerational Ambivalence Among Chinese Rural Elderly Population (N = 4,396)

The results of the Model 2 show that controlling for children’s characteristics reduced the mean ambivalence scores (M = −4.27). Parents reported a significantly higher level of ambivalence toward sons than toward daughters. Parents were less ambivalent toward children who were professionals or entrepreneurs than toward children who were farmers. Children’s urban hukou status was also associated with a significantly lower level of ambivalence among parents. Parents also reported greater ambivalence toward children who lived closer and who had more frequent contact with them. Children’s reliance on parents for either monetary or instrumental support was positively associated with parents’ ambivalent feelings. The inclusion of children-level variables accounted for some of the variance in parents’ ambivalent feelings, but there remained variance unaccounted for, evident by the significant estimates of the residuals.

The results of the Model 3 show that none of the parents’ characteristics were associated with their ambivalent feelings. Unaccounted variance remained after adding parents’ attributes.

We also carried out separate multilevel regression analyses to predict positive and negative feelings, respectively, using the same predictors. The results showed that parents perceived sons more negatively and less positively compared with daughters and perceived children with urban hukou more positively and less negatively compared with children with rural hukou. Parents had both, more positive and negative feelings toward children who received help with childcare. The amount of net monetary transfer from parents to children was associated with greater negative feelings but not less positive feelings toward children.

Given the strong son preference and higher expectations on sons in China, we further tested whether predictors of ambivalence have a more significant impact on parent–son relations than on parent–daughter relations. We did so by adding interaction terms between children’s gender and all the other significant predictors of ambivalence in the regression (i.e., children’s occupation, hukou status, geographic distance from parents, frequency of contact with parents, and support from parents). The results identified only one significant interaction between gender and hukou. For the purpose of presentation, we only listed this interaction in Model 4 (Table 2). The negative association between the interaction term and ambivalence suggests that, the influence of children’s urban hukou status in reducing parents’ ambivalence is even stronger if the child is a son. To depict this interaction graphically, we calculated predicted ambivalence according to children’s gender and hukou status. All other covariates were held constant at their sample means in order to provide a picture of an “average” parent and an “average” child. As shown in Figure 1, parents had the highest level of ambivalence toward sons with rural hukou status, followed by daughters with rural hukou, sons with urban hukou, and daughters with urban hukou. Sons’ urban hukou status reduced parents’ ambivalence score by 1.03 points, whereas daughters’ urban hukou reduced the score by 0.71 points.

Figure 1.
Predicted level of intergenerational ambivalence among Chinese rural parents by children’s gender and children’s hukou status. Note. Analysis of variance (ANOVA) tests indicate significant group differences between sons with rural hukou ...

To better understand the mechanism through which hukou status affected parents’ feelings of ambivalence, we compared children who had urban hukou with those who had rural hukou in their geographic proximity, frequency of contact, and support exchanges with the parents. The purpose of the test is to see whether the influence of the hukou status on intergenerational ambivalence also reflects changes in other aspects of family dynamics that are associated with children’s hukou status. Given the gender-patterned family relations in rural China, we conducted separate analyses for sons and daughters.

The results of the analyses are shown in Table 3. As shown, regardless of children’s gender, compared with children who had rural hukou, those who had urban hukou lived further from parents and provided more monetary support to them. For sons, having urban hukou status was also associated with less personal contact with parents and less childcare assistance from the parents. The findings suggest that the greater ambivalence toward children with urban hukou was partially due to the greater distance, less intergenerational contact, and greater independence of these children. All these factors were associated with feelings of ambivalence among parents (Table 2, Model 2). Beyond the indirect effects, children’s hukou status was still directly related to parents’ ambivalent feelings, evidenced by the Model 2 in the regressions, in which children’s hukou remained a significant predictor even controlling for the factors examined previously.

Table 3.
Comparisons of Intergenerational Relations Between Children With Urban Hukou and Children with Rural Hukou by Children’s Gender

Finally, given that the age of the children in this study ranged from 15 to 73 years and that some variables (e.g., help with childcare from parents) may not be applicable to younger and older children, we repeated the regression analyses in three subsamples of children: those aged between 18 and 40, those aged between 18 and 50, and those aged between 18 and 60. The analyses yielded identical predictors of intergenerational ambivalence, suggesting that the findings in this study are robust among children of different age groups.


Existing literature on intergenerational ambivalence is limited in its reliance on samples from either United States or Europe. Particularly lacking are theoretically oriented studies that examine the notion of sociological ambivalence with testable predictions in societies other than the West. This study extends the efforts of previous work by studying the sources of intergenerational ambivalence among rural Chinese elders.

Inconsistent with the Western literature, which reports a greater sense of ambivalence among dyads with women than with men, the results of this study showed that Chinese rural parents had greater ambivalence toward sons than toward daughters. Strong son preference in rural China may contribute to this finding because the greater expectations on sons may yield not only affectionate, devoting, but also conflictual, disappointing, and burdensome relationships among parent–son ties, resulting in mixed feelings among parents. The strong son preference in China also affects intergenerational ambivalence indirectly through distinctive gender-patterned family interactions. Compared with daughters, sons in this study lived closer to their parents and had more in-person contact with them. We found that although parents perceived daughters more positively compared with sons, functionally, sons still largely assumed the socially assigned roles of old-age support, evident by the greater monetary and instrumental support provided by sons to the parents. Such relationships are reciprocal because sons also received more support, both monetarily and instrumentally, from parents. The great interdependence between sons and parents in China is likely to contribute to ambivalent feelings across generations. It is also possible that daughters-in-law may play a role in the pattern observed. In China, daughters-in-law often report sensitive and sometimes tense relationships with in-laws, in particular, mothers-in-law (Cong & Silverstein, 2008; Gallin, 1994). This in turn may affect the relationships between older parents and the sons. This assumption invites further examinations in future studies. Overall, the strong son preference in rural China shapes almost all aspects of parent–child relationships. Although sons play a crucial role in maintaining eldercare in rural China, they can be an important source of emotional complexity in intergenerational relations.

Our findings also support the second hypothesis, showing that children’s higher SES in general and having an urban hukou status in particular, were associated with parents’ lower levels of ambivalence. We argue that, unique to the Chinese society, having an urban hukou status, particularly among rural residents, encompasses dimensions of socioeconomic resources that are not fully captured by indicators of social stratification commonly used in Western studies. As shown, neither education nor financial conditions were significant predictors in this study. An urban hukou status provides rural residents with more secure and prestigious jobs and, perhaps more importantly, with substantial social benefits such as housing, medical care, and pensions. Although those with rural hukou can move freely to cities, they may not be able to bring their families or to settle down in the more prosperous urban areas. This broader institutional context is likely to influence parental expectations in rural China in the way that having urban hukou may be viewed as a more desired life circumstance for adult children. Children’s failure to do so thus generated parents’ ambivalent feelings, reflecting conflicts between their expectations for children’s success and a lack of social structure for such upward mobility. The influence of hukou status on ambivalence was particularly strong among parent–son ties because rural parents normally have higher expectations on sons for success. This finding again demonstrates how parental expectations and intergenerational relationships are influenced by the larger social and economic context.

The findings of this study support the third hypothesis that ambivalence increases under conditions of intergenerational dependence. Net monetary transfers from parents to children were linked to parents’ greater ambivalence. The finding is not surprising given the relatively poor financial status of the Chinese rural elderly population. But when we rerun the analysis using the absolute amount, instead of the net value, of the monetary support from parents to children as a predictor, the variable lost its significance. The findings suggest that, it is the sense of dependence not the sense of giving that generated ambivalent feelings among older parents. Even among parents have strong altruistic motivation and who are willing to help the children in need, children’s reliance on them may impair the reciprocity and the balance of dependence/independence across generations, leading to parents’ mixed feelings.

Helping children with childcare was also associated with older parents’ greater sense of ambivalence. Interestingly, such a grandparenting role was associated with both more positive and negative feelings toward the adult children, showing that the positive and negative sentiments are not mutually exclusive. The grandparenting role may provide older adults with a sense of fulfillment and may enhance the emotional cohesion between older parents and their offspring. But older parents may feel distress and exhausted at the same time due to the greater responsibilities associated with the role, a lack of time of their own, or potential conflict in other roles such as farm work and livelihood maintenance. Disagreement on childcare issues across generations may have also produced such ambivalence. Although previous studies have reported the psychological benefit of the grandparenting role among rural Chinese elderly population (Silverstein, Cong, & Li, 2007), the findings of this study revealed that such a role can also be a source of mixed feelings, leading to potential problems in cross-generational interactions.

This study has several limitations. The cross-sectional design of this study makes it difficult to establish causal relationships. For instance, the associations between parents’ support to children and their ambivalence may be explained as that parents who were more ambivalent toward children had different patterns of exchanges with them. Second, the information presented in this study was collected from older adults only. The information on children such as their economic conditions and monetary support exchanges may not be completely accurate as well. The assessment of intergenerational ambivalence also cannot speak for the feelings of adult children. Collecting information from both parents and children in future studies would yield a more accurate picture of family dynamics. This study is also limited in its generalizability. China is a huge country with significant variations in socioeconomic contexts across regions. The finding from this study may not be applied to older adults in urban areas who have more resources and to whom having an urban hukou may not be a concern or an involving expectation.

Despite these limitations, this study expands previous research of intergenerational ambivalence by testing the concept in a traditional Eastern society. To the best of our knowledge, this is the first time the concept was investigated among Asian elders using a quantitative approach. The findings identified both similar (intergenerational dependence) and distinctive predictors (sons, children’s hukou status) of intergenerational ambivalence among Chinese elderly population compared with their Western counterparts. It is also the first time the concept of sociological ambivalence has been formally tested by linking the social conditions and structures at the macro level with individual experience at the micro level. Moving beyond the common explanations of sociological ambivalence as a result of competing normative expectations, our study demonstrates that individual feelings of ambivalence are culturally structured and are determined by the complex local environment. The heterogeneity in intergenerational relations in rural China is influenced by broader social structures that not only define normative expectations of each generation (greater reliance on sons), shape family interactions (rural–urban migration and skip-generation households in China), but also constrain family members’ ability in exerting agency (structural barrier of the hukou system). As stated by Hillcoat-Nallétamby and Phillips (2011), ambivalence “cannot be reduced to an individual experience, disembodied from the wider web of interdependent social relationships to which it belongs and the social environments within which it is embedded” (p. 212).

As societies continue to evolve, changes in social structure and traditional values may keep reshaping parent–child relationships. The effect of these changes is particularly strong in families in Asia. Family researchers who have mainly focused on family systems in North American and European countries need to include more developing countries and transitional economies in their theorization. More family studies need to be conducted to understand the local context of family relations and how it influences the needs, expectations, and capacities of family members.


This work was supported by grants from the Fogarty International Center of the National Institutes of Health (R03TW01060-01) and the U.S.–China Institute and School of Social Work at University of Southern California.


The authors would like to thank Shuzhuo Li for his leadership as the Co-director of the larger research project and Dongmei Zuo for her contribution to data collection. M. Guo planned the study, conducted the data analysis, and wrote the paper; I. Chi helped plan the study and contributed to revising the paper; and M. Silverstein helped plan the study.


  • Albright J. J. (2007). Estimating multilevel models using SPSS, Stata, and SAS Retrieved from
  • Bengtson V. L., Giarrusso R., Mabry J. B., Silverstein M. (2002). Solidarity, conflict, and ambivalence: Complementary or competing perspectives on intergenerational relationships? Journal of Marriage and Family, 64, 568–576doi:10.1111/j.1741-3737.2002.00568.x.
  • Bengtson V. L., Rosenthal C. J., Burton L. (1996). Paradoxes of families and aging. In Binstock R. H., George L. K, editors. (Eds.), Handbook of aging and the social sciences, (4th ed., pp. 253–282). New York: Academic Press;
  • Birditt K. S., Fingerman K. L., Zarit S. H. (2010). Adult children’s problems and successes: Implications for intergenerational ambivalence. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 65, 145–153 doi:10.1093/geronb/gbp125. [PMC free article] [PubMed]
  • Birditt K. S., Jackey L. M. M., Antonucci T. C. (2009). Longitudinal patterns of negative relationship quality across adulthood. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 64, 55–64doi:10.1093/geronb/gbn031. [PMC free article] [PubMed]
  • Chen F., Liu G. (2009). Population aging in China. In Uhlenberg P, editor. (Ed.), International handbook of population aging, (pp. 157–172). New York: Springer;
  • Cong Z., Silverstein M. (2008). Intergenerational support and depression among elders in rural China: Do daughters-in-laws matter? Journal of Marriage and Family, 70, 599–612doi:10.1111/j. 1741-3737.2008.00508.x.
  • Connidis I. A., McMullin J. A. (2002a). Ambivalence, family ties, and doing sociology. Journal of Marriage and Family, 64, 594–601doi:10.1111/j.1741-3737.2002.00594.x.
  • Connidis I. A., McMullin J. A. (2002b). Sociological ambivalence and family ties: A critical perspective. Journal of Marriage and F, amily, 64, 558–567doi:10.1111/j.1741-3737.2002.00558.x.
  • de Leon M., Glass C. (2004). The role of social and personal resources in ethnic disparities in late-life health. In Anderson N. B., Bulatao R. A., Cohen B, editors. (Eds.), Critical perspectives on racial and ethnic differences in health in late life (Vol. 1, pp. 353–405). Washington, DC: The National Academies Press;
  • Fan C. C. (2002). The elite, the natives, and the outsiders: Migration and labor market segmentation in urban China. Annals of the Association of American Geographers, 27, 103–124doi:10.1111/1467-8306.00282.
  • Fingerman K. L. (1998). Tight lips?: Aging mothers’ and adult daughters’ responses to interpersonal tensions in their relationships. Personal Relationships, 5, 121–138doi:10.1111/j.1475-6811.1998.tb00163.x.
  • Fingerman K. L. (2001). Aging mothers and their adult daughters: A study in mixed emotions. New York: Springer Publishing Company;
  • Fingerman K. L., Chen P. C., Hay E., Cichy K. E., Lefkowitz E. S. (2006). Ambivalent reactions in the parent and offspring relationship. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 61, 152–160doi:10.1093/geronb/61.3.P152. [PubMed]
  • Fingerman K. L., Hay E. L. (2004). Intergenerational ambivalence in the context of the larger social network. In Pillemer K., Lüscher K, editors. (Eds.), Intergenerational ambivalences: New perspectives on parent-child relations in later life. (pp. 133–152). New York: Emerald Group Publishing;
  • Fingerman K. L., Pitzer L., Lefkowitz E. S., Birditt K. S., Mroczek D. (2008). Ambivalent relationship qualities between adults and their parents: Implications for the well-being of both parties. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 63, 362–371doi:10.1093/geronb/63.6.P362. [PMC free article] [PubMed]
  • Gallin R. S. (1994). The intersection of class and age: Mother-in-law/daughter-in-law relations in rural Taiwan. Journal of Cross-Cultural Gerontology, 9, 127–140doi:10.1007/BF00972145. [PubMed]
  • Grzywacz J. G., Quandt S. A., Early J., Tapia J., Graham C. N., Arcury T. A. (2006). Leaving family for work: Ambivalence and mental health among Mexican migrant farmworker men. Journal of Immigrant and Minority Health, /, Center for Minority Public Health, 8, 85–97doi:10.1007/s10903-006-6344-7. [PubMed]
  • Guo M., Chi I., Silverstein M. (2012). The structure of intergenerational relations in rural China: A latent class analysis. Journal of Marriage and Family, 74, 1114–1128.
  • Guo M., Chow N. W. S., Palinkas L. A. (2011). Circular migration and life course of female domestic workers in Beijing. Asian Population Studies, 7 , 51–67doi:10.1080/17441730.2011.544905.
  • Ha J. H., Ingersoll-Dayton B. (2008). The effect of widowhood on intergenerational ambivalence. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 63, 49–58doi:10.1093/geronb/63.1.S49. [PMC free article] [PubMed]
  • Hillcoat-Nallétamby S., Phillips J. E. (2011). Sociological ambivalence revisited. Sociology, 45, 202–217doi:10.1177/0038038510394018.
  • Ingersoll-Dayton B., Dunkle R. E., Chadiha L., Lawrence-Jacobson A., Li L., Weir E., Satorius J. (2011). Intergenerational ambivalence: aging mothers whose adult daughters are mentally ill. Families in Society: The Journal of Contemporary Human Services, 92, 114–119doi:10.1606/1044-3894.4077. [PMC free article] [PubMed]
  • Jingzhong Y., Lu P. (2011). Differentiated childhoods: Impacts of rural labor migration on left-behind children in China. The Journal of Peasant Studies, 38, 355–377doi:10.1080/03066150.2011. 559012. [PubMed]
  • Krause N., Rook K. S. (2003). Negative interaction in late life:. Issues in the stability and generalizability of conflict across relationships. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 58, 88–99doi:10.1093/geronb/58.2.P88. [PMC free article] [PubMed]
  • Letiecq B. L., Bailey S. J., Dahlen P., Hayslip B., Kaminski P. (2008). Ambivalence and coping among custodial grandparents. In Hayslip B., Kaminski P. L, editors. (Eds.), Parenting the custodial grandchild. Implications for clinical practice (pp. 3–16). New York: Springer Publishing Company;
  • Levitzki N. (2009). Parenting of adult children in an Israeli sample: Parents are always parents. Journal of Family Psychology, 23, 226–235doi:10.1037/a0015218. [PubMed]
  • Lewis D. C. (2008). Types, meanings and ambivalence in intergenerational exchange among Cambodian refugee families in the United States. Aging & Society, 18, 693–715doi:10.1017/S0144686X08007034.
  • Lowenstein A. (2007). Solidarity-conflict and ambivalence:. Testing two conceptual frameworks and their impact on quality of life for older family members. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 62, 100–107doi:10.1093/geronb/62.2.S100. [PubMed]
  • Lüscher K. (2002). Intergenerational ambivalence: Further steps in theory and research. Journal of Marriage and Family, 64, 585–593doi:10.1111/j.1741-3737.2002.00585.x.
  • Lüscher K., Pillemer K. (1998). Intergenerational ambivalence: A new approach to the study of parent-child relations in later life. Journal of Marriage and the Family, 60, 413–425doi:10.2307/353858.
  • Merton R. K., Barber E. (1963). Sociological ambivalence. In Tiryakia E. A, editor. (Ed.), Sociological theory: Values and sociocultural change: Essays in honor of Pitirim A. Sorokin (pp. 91–120). New York: Free Press;
  • National Bureau of Statistics of China (2011). China statistical yearbook 2011. Beijing: China: Statistics Press;
  • Peters C. L., Hooker K., Zvonkovic A. M. (2006). Older parents’ perceptions of ambivalence in relationships with their children. Family Relations, 55, 539–551doi:10.1111/j.1741-3729.2006.00424.x.
  • Pillemer K., Suitor J. J. (2002). Explaining mothers’ ambivalence toward their adult children. Journal of Marriage and Family, 64, 602–613doi:10.1111/j.1741-3737.2002.00602.x.
  • Silverstein M., Cong Z., Li S. (2007). Grandparents who care for their grandchildren in rural China: Benefactors and beneficiaries. In Powell J. L., Cook I. G, editors. (Eds.), New perspectives on China and aging (pp. 49–72). New York: Nova Publishers;
  • Smelser N. J. (1998). The rational and the ambivalent in the social sciences. American Sociological Review, 63, 1–16doi:10.2307/2657473.
  • StataCorp (2005). Stata statistical software: Release 9. College Station, TX: StataCorp LP;
  • Thornton A., Fricke T., Yang L. S. (1994). Theoretical mechanisms of family change. In Thornton A., Lin H. S, editors. (Eds.), Social change and the family in Taiwan (pp. 88–115). Chicago, IL: University Of Chicago Press;
  • Traphagan J. W. (2010). Intergenerational ambivalence, power, and perceptions of elder suicide in rural Japan. Journal of Intergenerational Relationships, 8, 21–37doi:10.1080/15350770903520643.
  • Treas J., Mazumdar S. (2002). Older people in America’s immigrant families: Dilemmas of dependence, integration, and isolation. Journal of Aging Studies, , 16, 243–258doi:10.1016/S0890-4065(02)00048-8.
  • Whyte M., Qin X. (2003). Support for aging parents from daughters versus sons. In Whyte M, editor. (Ed.), China’s revolutions and intergenerational relations, (pp. 167–196). Ann Arbor, MI: Center for Chinese Studies, the University of Michigan;
  • Willson A. E., Shuey K. M., Elder G. H., Jr (2003). Ambivalence in the relationship of adult children to aging parents and in-laws. Journal of Marriage and Family, 65, 1055–1072doi:10.1111/j. 1741-3737.2003.01055.x.
  • Willson A. E., Shuey K. M., Elder G. H., Jr., Wickrama K. A. S. (2006). Ambivalence in mother-adult child relations: A dyadic analysis. Social Psychology Quarterly, , 69, 235–252doi:10.1177/ 019027250606900302.
  • Xu Y. (2001). Family support for old people in rural China. Social Policy and Administration, 35, 307–320doi:10.1111/1467-9515.00235.
  • Zhan H. J., Montgomery R. J. V. (2003). Gender and elder care in China: The influence of filial piety and structural constraints. Gender and Society, 17, 209–229doi:10.1177/0891243202250734.

Articles from The Journals of Gerontology Series B: Psychological Sciences and Social Sciences are provided here courtesy of Oxford University Press