Introduction
With the decline of fertility and mortality along with the increase in life expectancy, China is ageing rapidly. Technically, adult children and their elderly parents tend to have a longer shared life time nowadays than in the past, thus intergenerational relations and other issues related to elderly persons have received growing attention (Luescher and Pillemer, Reference Luescher and Pillemer1998; Lowenstein, Reference Lowenstein2007). To explain parent–child interaction, especially intergenerational family cohesion, previous studies have developed the intergenerational solidarity theory, which refers to multi-dimensional elements including affection, association, consensus, resource sharing and strength of familism norms (Bengtson and Roberts, Reference Bengtson and Roberts1991). However, the solidarity theory over-emphasises the positive, harmonious and consensus aspects of intergenerational relations while downplaying negative or conflicted aspects of family dynamics (Luescher and Pillemer, Reference Luescher and Pillemer1998; Park, Reference Park2014). To explain the complexity of family ties, some scholars proposed the concept of intergenerational ambivalence in the form of conflicting emotions and cognition, emphasising the co-existence of harmony and conflict in intergenerational relations (Luescher and Pillemer, Reference Luescher and Pillemer1998; Connidis and Mcmullin, Reference Connidis and Mcmullin2002). Existing evidence reveals that parents and their adult children often have mixed feelings towards each other, suggesting that intergenerational relationships involve the ambivalence of positive and negative feelings simultaneously (Pillemer and Suitor, Reference Pillemer and Suitor2002; Birditt et al., Reference Birditt, Miller, Fingerman and Lefkowitz2009, Reference Birditt, Fingerman and Zarit2010).
Previous studies have devoted much research to describing the sources or outcomes of ambivalence (Willson et al., Reference Willson, Shuey, Elder and Wickrama2006; Birditt et al., Reference Birditt, Fingerman and Zarit2010; Kiecolt et al., Reference Kiecolt, Blieszner and Savla2011; Tighe et al., Reference Tighe, Birditt and Antonucci2016), but few studies have examined the relationship between traditional culture and intergenerational ambivalence in developing countries. To fill this gap in the literature, the present study examines the relationships between filial piety and intergenerational ambivalence among mother–adult child dyads in rural China using the intergenerational ambivalence theory. We focus on intergenerational ambivalence of elderly mothers and their adult children for several reasons. First, as the primary kinship keepers who maintain family relationships and promote affective interactions in the family (Leach and Braithwaite, Reference Leach and Braithwaite1996), mothers experience greater ambivalence compared to fathers (Tighe et al., Reference Tighe, Birditt and Antonucci2016). Second, as women's life expectancy generally is longer compared to men's (Le et al., Reference Le, Ren, Shen, Li and Zhang2015), elderly mothers may experience a longer lifetime with adult children. Hence, a study on the intergenerational relationships between elderly mothers and adult children has conceptual and practical implications.
Literature review
Intergenerational ambivalence
The concept of intergenerational ambivalence is described as ‘contradictions in relationships between parents and adult offspring that cannot be reconciled’ (Luescher and Pillemer, Reference Luescher and Pillemer1998: 416), which emphasises mixed feelings (i.e. positive and negative feelings) in the same child–parent relationship (Luescher and Pillemer, Reference Luescher and Pillemer1998; Pillemer and Suitor, Reference Pillemer and Suitor2002). To examine contradictory feelings within the same relationship, the most popular measurement of ambivalence is the ‘Griffin measure’, developed by Thompson et al. (Reference Thompson, Zanna, Griffin, Petty and Krosnick1995), which asked respondents to rate their positive and negative attitudes towards the other party in the relationship and creates a numeric value representing the balance between these two feelings. Most studies only focus on combined ambivalence of both positive and negative feelings simultaneously using the Griffin measure (Birditt et al., Reference Birditt, Fingerman and Zarit2010; Tighe et al., Reference Tighe, Birditt and Antonucci2016), while other studies also examine these two components separately (Willson et al., Reference Willson, Shuey, Elder and Wickrama2006; Ngai, Reference Ngai2014).
Previous studies have proposed three aspects of parent–child relations in later life that are likely to generate ambivalence. Firstly, ambivalence is associated with dependence and autonomy. Parent's and children's expectations of mutual help conflict with the desire for freedom from the parent–child relationship (Luescher and Pillemer, Reference Luescher and Pillemer1998; Pillemer and Suitor, Reference Pillemer, Suitor, Silverstien, Giarruso and Bengtson2005). For example, elderly mothers might struggle with the choices of either helping their daughters take care of their grandchildren or becoming community volunteers (Cohler and Grunebaum, Reference Cohler and Grunebaum1981). Moreover, some scholars have used employment, education and marital status to assess adult children's level of dependence and autonomy, finding that elderly parents might feel more ambivalent when their adult children are not sufficiently independent (Fingerman et al., Reference Fingerman, Chen, Hay, Cichy and Lefkowitz2006; Birditt et al., Reference Birditt, Fingerman and Zarit2010). In addition, ambivalence is related to incompatible norms regarding intergenerational relations. These norms usually involve filial responsibility, commitment to assisting another generation, obligation to kin and reciprocity norm (George, Reference George, Pillemer and Wolf1986; Luescher and Pillemer, Reference Luescher and Pillemer1998). For example, based on the reciprocity norm, disabled parents may expect support from their children but they also feel guilty that they cannot reciprocate (George, Reference George, Pillemer and Wolf1986). Finally, ambivalence is related to solidarity factors, such as functional support, close proximity and contact, and is also accompanied by opposites, such as direct conflict and struggle for independence (George, Reference George, Pillemer and Wolf1986; Luescher and Pillemer, Reference Luescher and Pillemer1998). Ambivalent feelings are prevalent between abused elderly parents and their abusive children, partly because the parents are physically impaired and typically require help from the abusive children (Wolf and Pillemer, Reference Wolf and Pillemer1989).
Other characteristics, such as age, marital status, education and health also play a role in intergenerational ambivalence (Pillemer and Suitor, Reference Pillemer and Suitor2002; Kiecolt et al., Reference Kiecolt, Blieszner and Savla2011). For instance, age is a significant factor in ambivalence, e.g. compared with emerging adults (ages 18–25 years), young adults (ages 26–29 years) and adolescents (ages 13–17 years) report greater ambivalence to fathers (Tighe et al., Reference Tighe, Birditt and Antonucci2016). Unmarried children or those with lower education levels demonstrate greater ambivalence towards their parents (Pillemer and Suitor, Reference Pillemer, Suitor, Silverstien, Giarruso and Bengtson2005; Pillemer et al., Reference Pillemer, Suitor, Mock, Sabir, Pardo and Sechrist2007). Moreover, parent characteristics, such as poor health or neurotic personality, are associated with more ambivalence from children (Pillemer et al., Reference Pillemer, Suitor, Mock, Sabir, Pardo and Sechrist2007; Kiecolt et al., Reference Kiecolt, Blieszner and Savla2011).
However, most previous studies have utilised the intergenerational ambivalence theory to unravel the complex family bonds in developed countries, such as the United States of America (USA), Germany and Japan (Pillemer and Suitor, Reference Pillemer and Suitor2002; Traphagan, Reference Traphagan2010; Prinzen, Reference Prinzen2012), while largely ignoring intergenerational ambivalence in developing countries. Thus, little is known regarding the relationship between filial piety and intergenerational ambivalence. Different from other counties, intergenerational relations in China are deeply influenced by Confucian culture, which emphasises filial piety and family cohesion. Thus, this study considers cultural context in explaining intergenerational ambivalence in rural China, a traditional society that is undergoing modernisation and social transformation including cultural changes related to filial piety.
Filial piety in rural China
As a core virtue in Confucianism, filial piety is the guiding principle in Chinese society that specifies children's behaviours towards their parents (Chow, Reference Chow, Chi, Chappel and Lubben2001; Cheng and Chan, Reference Cheng and Chan2006). Filial piety involves practising filial respect of and care for parents (Khalaila, Reference Khalaila2010; Khalaila and Litwin, Reference Khalaila and Litwin2011) and exemplifies both attitudes and behaviours (Luo and Zhan, Reference Luo and Zhan2012). It covers a wide range of filial attributes, including love, respect, obedience and loyalty, which are demonstrated by providing sufficient financial, physical and emotional support to elderly parents, maintaining contact or living with parents, as well as showing respect and obedience to parents, honouring or promoting family prestige, continuing the family line, etc. (Ho, Reference Ho1994; Cheng and Chan, Reference Cheng and Chan2006; Luo and Zhan, Reference Luo and Zhan2012). Further, Sung (Reference Sung1995) suggests that these multiple indicators include both behavioural and emotional components. Filial piety always involves both children and their parents (Hanson et al., Reference Hanson, Sauer and Seelbach1983; Cicirelli, Reference Cicirelli1993). Some studies have focused on the behaviours or attitudes that adult children use to reflect their value of filial piety, such as taking care of elderly parents, showing respect, and providing room and financial support (Ho, Reference Ho1994; Sung, Reference Sung1998; Ng et al., Reference Ng, Loong, Liu and Weatherall2000). Other studies have reported parents’ perceptions of their adult children's filial piety regarding care and respect (Lee et al., Reference Lee, Netzer and Coward1994; Lee and Hong-Kin, Reference Lee and Hong-Kin2005; Cheng and Chan, Reference Cheng and Chan2006; Luo and Zhan, Reference Luo and Zhan2012). For example, Luo and Zhan (Reference Luo and Zhan2012) described normative solidarity as filial piety, which referred to elderly parents’ evaluation of their adult children's filial piety.
In light of the previous literature, this study considers filial piety from the perspective of elderly mothers and from two aspects. The first aspect includes children's specific filial behaviours and attitudes as reported by their elderly mothers, including financial support, daily care, pleasing, contact, obedience, respect and living distance. The majority of elderly parents in rural China still receive a substantial amount of financial, instrumental and emotional support from their children (Pei and Pillai, Reference Pei and Pillai1999; Lei, Reference Lei2013), and elderly parents co-reside with children in some regions (Xie and Zhu, Reference Xie and Zhu2006). Adult children who migrate still maintain frequent contact with their parents in rural areas (Luo and Zhan, Reference Luo and Zhan2012). The other aspect involves normative solidarity and elderly parents’ filial expectations, which are also treated as filial piety (Luo and Zhan, Reference Luo and Zhan2012). To specify, for this study, this indicator of filial piety refers to elderly mothers’ overall evaluation of their children's specific filial behaviours and attitudes, i.e. how they evaluate their children's degree of filial piety. According to the theory of intergenerational solidarity, rating the strength of filial piety is a common way to measure the normative solidarity (Bengtson and Roberts, Reference Bengtson and Roberts1991). In China, the universal term for filial piety (xiao) is used to evaluate a personal trait or attribute, which also implies traditional cultural expectations and parents’ filial expectations (Luo and Zhan, Reference Luo and Zhan2012). If elderly parents have higher evaluations of adult children's filial piety, this may signify that their children's filial behaviours and attitudes are in accordance with traditional culture and parents’ filial expectations.
Implications of filial piety for intergenerational ambivalence
Few studies have focused on the correlation between filial piety and intergenerational ambivalence, but many have involved the relationship between intergenerational solidarity and intergenerational ambivalence (Birditt et al., Reference Birditt, Miller, Fingerman and Lefkowitz2009, Reference Birditt, Fingerman and Zarit2010). Some components of solidarity and filial behaviours overlap when comparing intergenerational solidarity with filial piety. Similar to intergenerational solidarity, which is a known contributor to ambivalence, filial piety may also generate ambivalence. According to the model of intergenerational solidarity (Bengtson and Roberts, Reference Bengtson and Roberts1991), financial support, emotional support and daily care are aspects of functional solidarity, and living distance between generations is part of structural solidarity, contact is part of associational solidarity and the rating of strength of filial piety is a component of normative solidarity. Most components of solidarity are consistent with the characteristics of filial piety. However, solidarity emphasises mutual reciprocity between generations, while filial piety focuses on the support from children to parents. Many studies have analysed the relationship between downward support (from parents to children) and ambivalence (Pillemer and Suitor, Reference Pillemer and Suitor2002; Pillemer et al., Reference Pillemer, Suitor, Mock, Sabir, Pardo and Sechrist2007; Birditt et al., Reference Birditt, Fingerman and Zarit2010) as a reflection of children's independence, which is significantly related to parental ambivalence (Birditt et al., Reference Birditt, Fingerman and Zarit2010). However, far fewer studies have analysed the relationship between upward support (from children to parents) and ambivalence (Willson et al., Reference Willson, Shuey, Elder and Wickrama2006; Pei and Cong, Reference Pei and Cong2016).
In terms of specific filial behaviours and attitudes, previous studies have found that children living with parents can reduce parental ambivalence in the USA (Kiecolt et al., Reference Kiecolt, Blieszner and Savla2011), while other studies have found that living distance is not related to mothers’ ambivalence in the USA (Pillemer and Suitor, Reference Pillemer and Suitor2002; Willson et al., Reference Willson, Shuey, Elder and Wickrama2006). Some US studies have found that emotional support from adult children does not significantly affect parental ambivalence (Willson et al., Reference Willson, Shuey, Elder and Wickrama2006; Kiecolt et al., Reference Kiecolt, Blieszner and Savla2011). Additionally, children's daily care has been found to be significantly related to mothers’ combined ambivalence and positive feelings, but not related to mothers’ negative feelings (Willson et al., Reference Willson, Shuey, Elder and Wickrama2006). Frequent contact with elderly parents can reduce Mexican-American elderly parents’ ambivalence, but financial support to parents does not significantly affect ambivalence (Pei and Cong, Reference Pei and Cong2016). These above studies are focused on Western countries, and the same filial behaviours and attitudes may not be valid in China, because traditional Chinese culture has stronger emphasis on filial piety.
Furthermore, the relationship between the overall evaluation by elderly mothers of their adult children's filial piety and intergenerational ambivalence remains unclear. Indicators of elderly mothers’ overall evaluation of their adult children's filial piety involve normative solidarity and filial expectations, and ambivalence theory states that incompatible norms can easily lead to ambivalence. Previous studies have found that parents always hold higher filial expectations than young people do (Harwood et al., Reference Harwood, Giles, McCann, Cai, Somera, Ng and Noels2001; Luo and Zhan, Reference Luo and Zhan2012), which could lead to inconsistency between parents’ filial expectations and the perceived filial piety of children, resulting in ambivalence. A study in Hong Kong has found that children who value filial piety (xiao) do not generate ambivalence in their parents, but rather more positive feelings and increased maintenance of intergenerational relationship with their parents (Ngai, Reference Ngai2014). Additionally, if elderly mothers evaluate their children filial (xiao), there is need to determine whether it will increase mothers’ positive feeling without generating their ambivalence to children.
Thus, the present study aims to examine, from the perspective of elderly mothers, the relationship between components of filial piety (children's filial behaviours and attitudes as well as elderly mothers’ overall evaluation of children's filial piety) and the ambivalence of elderly mothers towards their adult children (including positive feelings, negative feelings and combined ambivalence).
Methods
Data and sample
A community-based and cross-sectional survey was conducted in rural areas in four towns in S County (a pseudonym) of Sichuan Province in China between January and February 2016. The survey was administered by the School of Sociology at Huazhong University of Science and Technology. This particular county was selected for the study because population ageing in this community was much higher than the national average. By the end of 2014, 15.5 per cent of the population of China was aged ⩾60 years, while the proportion in S County was 19.75 per cent. The high proportion of those aged 60 or older was due to two specific reasons: first, before the one-child policy was abolished in 2015 in China, compared with other provinces, Sichuan Province implemented a stricter one-child policy due to its larger population base; second, population out-migration from Sichuan, including S County, was common, and the out-migration of large proportions of the labour force exacerbated the process of population ageing in such areas. For example, S County had approximately 1.5 million residents, including half a million rural–urban migrants who worked outside this area. The proportion of out-migration population in S County was about 33 per cent, while according to the 2016 report on the development of China's floating population, the migration population was approximately 247 million by the end of 2015, accounting for 18 per cent of the total population in China.
We used combined multi-stage systematic sampling for the survey. In the first stage, we randomly selected four towns from the four cardinal directions (north, south, east and west), which were divided by local government according to economic status. In the second stage, we selected ten villages from each town using isometric sampling based on the geographical location. In the third stage, we extracted three village groups (the lowest administrative unit determined by the government) from each village according to isometric sampling and based on geographical location. In the last stage, we selected ten participants among each village group aged ⩾60 years with at least one child aged ⩾20 years to take the survey. If the elderly parent was widowed, we selected them as our participant, and if both of the elderly couple were alive, we only selected the women, resulting in a larger proportion of elderly women in our survey. Our sampling gave preference to elderly women due to their relatively weak position yet important family kinship keepers in rural areas. Elderly rural women are more likely than elderly men to have a higher poverty and illiteracy rate, higher widowhood rate, and lower social and family status (Zhang, Reference Zhang2009). Therefore, over-sampling women could shed light on the intergenerational ambivalence experienced by mothers in rural China and serve to empower elderly women in the family.
In our survey, we investigated 120 village groups from 40 villages in total, and the sample consists of 1,104 respondents (24.28% male, 75.72% female). A small percentage of participants (about 2.5%) refused to be surveyed, and there were a few vacant village groups (about ten village groups, most located in remote rural areas) due to population migration. We recruited 40 college students from a local town to conduct face-to-face structured interviews using local dialects. The average interview took approximately one hour. Participants provided demographic information about themselves and all of their children. As stated previously, we focused on elderly mothers. Excluding missing data, the sample consists of 802 female respondents with 2,203 adult children in the analysis. Participants had an average of 2.82 (standard deviation = 1.19) adult children aged >20 years (range 1–8). Approximately 80 per cent of elderly mothers ranged in age from 60 to 80 years, and nearly 70 per cent of children were aged between 36 and 50 years.
Measures
Ambivalence
We used an indirect approach to assess both positive and negative perceptions of the mother–children relationship to generate ambivalence scores for each child (Fingerman et al., Reference Fingerman, Chen, Hay, Cichy and Lefkowitz2006; Birditt et al., Reference Birditt, Fingerman and Zarit2010). For the positive component, we summed responses to two items: ‘How much does this child love and care for you?’ and ‘How much does this child understand you?’ For the negative component, we summed two questions: ‘How much does the child criticise you?’ and ‘How much does the child make demands on you?’ (Birditt et al., Reference Birditt, Fingerman and Zarit2010). Items were rated on a five-point scale from 1 = not at all to 5 = a great deal. In line with previous studies (Willson et al., Reference Willson, Shuey, Elder and Wickrama2006; Birditt et al., Reference Birditt, Fingerman and Zarit2010), we calculated indirect ambivalence scores using the Griffin measure formula (Thompson et al., Reference Thompson, Zanna, Griffin, Petty and Krosnick1995): [(positive + negative)/2 − |positive − negative|]. Higher scores indicated greater ambivalence. For the positive and negative feelings, we summed responses to two items consistent with a previous study (Willson et al., Reference Willson, Shuey, Elder and Wickrama2006).
Filial piety
One component of filial piety assessed specific filial behaviours and attitudes of adult children. Filial behaviours and attitudes were measured by seven items derived from the published literature (Gallois et al., Reference Gallois, Giles, Ota, Pierson, Ng, Lim and Harwood1997; Cheng and Chan, Reference Cheng and Chan2006), including financial support, daily care, pleasing, contact, obedience, respect and living distance. The financial support and daily care (i.e. cleaning, washing or cooking) domains were divided into two categories: 1 = support and 0 = no support. We replaced pleasing mothers with emotional support (Mangen et al., Reference Mangen, Bengtson and Landry1988); participants were asked the question, ‘Taking everything into consideration, how close do you feel to (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? Overall, how well do you and (this child) get along with each other?’ An answer of ‘Not at all close/not at all/not well’ was scored 1, ‘somewhat close/somewhat/somewhat well’ was scored 2 and ‘very close/very much/very well’ was scored 3. The aggregated score of the three items was used, with the higher score indicating more emotional support provided to mothers. Contact frequency was rated as 1 = irregularly, 2 = several times a year, 3 = once a week, 4 = several times a week, 5 = once a week and 6 = everyday. Obedience was measured by responses to the question, ‘To what extent would this child take your advice into consideration on important affairs such as marriage or migration?’, and items were rated from 1 = never taken to 5 = all taken. We used conflict frequency rather than level of respect as less conflict represents more respect or more filial piety to mothers. Participants were asked if they had any conflicts, strains or disagreements with the child in the past three months, which was scored as 1 = a great deal to 5 = not at all. We measured how close each child lived away from their mothers on a six-point Likert scale by living in 1 = other provinces, 2 = other cities but in Sichuan Province, 3 = another county in the city, 4 = county, 5 = town and 6 = the same village. Higher scores indicate closer proximity. The second component is the overall evaluation by elderly mothers of their children's specific filial behaviours and attitudes. Participants were asked to appraise each child's filial degree (xiao) from 1 = very unfilial to 5 = very filial. Control variables include (a) indicators reflecting the independence levels of children (Birditt et al., Reference Birditt, Fingerman and Zarit2010), such as marital status (1 = married, 0 = divorced/separated/unmarried), employment status (1 = non-agricultural, 0 = agricultural) and educational level (years of schooling); and (b) variables representing the content of solidarity, such as financial support and instrumental support from mothers to children over the past year (1 = support, 0 = no support). Instrumental support refers to whether the children received support with household work from their mothers, such as washing clothes, cleaning the house or cooking meals. Demographic variables include mothers’ marital status (1 = married, 0 = widowed/divorce), education level (1 = educated, 0 = illiterate), number of living children, income (log), self-rated health (1 = excellent, 5 = very poor) and age, as well as the children's gender (1 = daughter, 0 = son) and age.
Analytic strategy
Because most participants had more than one child, the 2,203 children who were the units of analysis were nested within the 802 elderly mothers who were interviewed. A two-level mixed model with random intercept was used to examine the effects of filial piety on ambivalence. The models include two levels: higher-level participant characteristics (e.g. age, income and number of living children) and lower-level child characteristics (e.g. filial behaviours, education level and gender). A two-level hierarchical linear model was evaluated separately for positive feelings, negative feelings and combined ambivalence. In addition to filial piety variables, models also include characteristics of both elderly mothers (e.g. education, number of living children, income and self-rated health) and adult children (e.g. gender, marital status, occupation and education level). A p-value of 0.01 or less is used to indicate statistical significance throughout the paper.
Results
Descriptive results
Table 1 presents the adult children's and elderly mothers’ characteristics for the analytic sample. The elderly mothers reported higher positive than negative feelings towards their children, leading to a moderate-level combined ambivalence score. Approximately 85 per cent of children provided financial support for their mothers, and mothers reported high emotional support and low levels of conflict with their children. The percentage of children providing daily care to mothers was low, as a number of adult children migrated to cities for work and the distance was not conducive to daily care. Different from the traditional form of filial piety emphasising absolute obedience, our results revealed a moderate degree of children's obedience to their mothers. When adult children were faced with important decision-making affairs, scoring from 1 ‘never taken mothers’ advice’ to 5 ‘taken all advice’, the average score was 2.41, which means that adult children partly took their mothers’ advice, and absolute obedience has no longer been practised by children in modern society. Moreover, the living distance between mothers and children has expanded, and the rural–urban out-migration has led to a decrease in daily care support provided to elderly mothers. In the overall evaluation of filial piety, mothers on average highly appraised their children's filial piety, from 1 ‘very unfilial’ to 5 ‘very filial’, and the average score was 3.94, indicating that although some elements of filial piety changed in rural China, such as absolute obedience no longer being accepted by children, reduced proportion of families living together and less frequent care to mothers, some aspects of filial piety remained the same, e.g. the majority of children still provided financial support and emotional support for their elderly mothers.
Table 1. Descriptive characteristics of mothers and their adult children
Note: SD: standard deviation.
Multi-level regression analysis
Table 2 presents the results of the multi-level linear regression analysis. For the children's filial behaviours and attitudes in the positive feelings model, mothers reported increased positive feelings when their children provided more financial support, more emotional support and less conflict. The model of negative feelings has revealed that more emotional support and less conflict from adult children were associated with fewer negative feelings from the mothers, but more obedience from the children was related to higher levels of negative feelings from the mothers. After integrating the positive and negative feelings, the model of combined ambivalence, similar to the results of the negative feeling model, has revealed that elderly mothers reported lower ambivalence when children provided more emotional support and less conflict. The more obedience children showed to their mothers was associated with greater ambivalence. Regarding children's filial degree, the higher level of children's filial piety was significantly related to higher levels of positive feelings and lower levels of negative feelings of elderly mothers, and children's higher filial degree was related to elderly mothers’ lower ambivalence. When the mothers identified their children as more filial, it showed that their children's behaviours and attitudes met their filial expectations, and that the norms between generations were compatible, resulting in less ambivalence for elderly mothers.
Table 2. Multi-level models examining the role of filial piety on ambivalence
Notes: Mothers: N = 802; mother–child pairs: N = 2,203. SE: standard error. Ref.: reference category.
Significance levels: *p < 0.01, **p < 0.001.
Ambivalence also varied according to the mothers’ and children's characteristics. Those elderly mothers who were married or had lower income typically reported higher levels of positive feelings. Participants also reported greater positive feelings towards their younger adult children. From the model of negative feelings, participants reported more negative feelings when they provided financial support to their adult children. In the combined ambivalence model, those elderly mothers with a higher income were associated with greater ambivalence.
Discussion
The ambivalence theory suggests that parents experience mixed feelings due to incompatible norms and solidarity with their adult children (Luescher and Pillemer, Reference Luescher and Pillemer1998; Pillemer and Suitor, Reference Pillemer and Suitor2002). In relation to China, norms and solidarity can be merged into filial piety. The present study has examined the relationship between filial piety and intergenerational ambivalence. Our results have shown that adult children's filial behaviours and attitudes were partly related to positive and negative feelings as well as combined ambivalence, but the strength and direction of the correlation were different.
We found that financial support is only related to elderly mothers’ positive feelings, rather than negative feelings and combined ambivalence. This is consistent with a prior study which also found financial support from children was unrelated to elderly parents’ ambivalence (Pei and Cong, Reference Pei and Cong2016). Unlike Western countries where most elderly persons receive an adequate pension, the Chinese rural elderly population does not receive sufficient pension support to meet their needs (Tao, Reference Tao2017). Thus, financial support from children to parents is expected in rural China, and many rural elderly parents rely on their children's financial support for daily needs (Pei and Pillai, Reference Pei and Pillai1999). Our findings concur that providing financial support to parents can be beneficial to intergenerational relations (Cong and Silverstein, Reference Cong and Silverstein2008; Kirchengast and Haslinger, Reference Kirchengast and Haslinger2015; Chen and Jordan, Reference Chen and Jordan2016).
Different from Western literature, which reports no relationship between children's emotional support and elderly mothers’ ambivalence (Willson et al., Reference Willson, Shuey, Elder and Wickrama2006; Kiecolt et al., Reference Kiecolt, Blieszner and Savla2011), this study showed that greater emotional support from children was associated with increased positive feelings and decreased negative feelings of Chinese rural elderly mothers, and was associated with a lower score of combined ambivalence. Emotional support can better promote the psychological wellbeing of elderly parents than financial support and daily care in China (Wang, Reference Wang2017), because emotional support reduces the loneliness and depression of elderly parents and increases their overall quality of life (Zhang and Du, Reference Zhang, Du, Cheng, Chi, Fung, Li and Woo2015; Wang, Reference Wang2017). Thus, more emotional support reduces ambivalence and negative feelings, which is in agreement with previous studies.
Contradictory to some findings in Western countries, which have shown daily care, contact frequency and living distance were related to parental ambivalence (Willson et al., Reference Willson, Shuey, Elder and Wickrama2006; Birditt et al., Reference Birditt, Fingerman and Zarit2010; Kiecolt et al., Reference Kiecolt, Blieszner and Savla2011), this study indicated these three components of filial piety were not associated with elderly mothers’ feelings of ambivalence. These results may be explained by the target place of our study, Sichuan, which is a province with a large out-migration population. As mentioned by prior research, the geographical distance between generations might block the delivery of daily care and contact from children (Zhang and Du, Reference Zhang, Du, Cheng, Chi, Fung, Li and Woo2015), and our sample also revealed such results. The separation between adult children and elderly mothers due to geographical distance may make the daily care and connection more appreciated, rather than leading to ambivalence in elderly mothers.
This study also has shown that elderly mothers reported greater positive feelings and fewer negative feelings, as well as fewer feelings of combined ambivalence, when elderly mothers had less conflict with children and appraised their children as more filial. Fewer conflicts between generations suggest compatible values, which is beneficial to intergenerational relationship quality. Not surprisingly, more conflicts are related to more negative feelings and a decline in overall relationship quality (Clarke et al., Reference Clarke, Preston, Raksin and Bengtson1999; Gaalen and Dykstra, Reference Gaalen and Dykstra2006). The overall evaluation by mothers of adult children's filial piety is a concept similar to normative solidarity (Luo and Zhan, Reference Luo and Zhan2012), and a mother who evaluates her children as more filial recognises her children's filial behaviours and attitudes, which is associated with less ambivalence. These findings suggest that compatible values and norms between generations are essential to reduce intergenerational ambivalence.
Unlike other filial behaviours and attitudes, obedience is positively correlated both with negative feelings and combined ambivalence. As a previous study revealed, there is no absolute obedience from children as it had been in the past, because the younger generation has achieved greater financial independence (Lee and Hong-Kin, Reference Lee and Hong-Kin2005), and most young people may no longer embrace obedience to parents as a way of demonstrating filial piety (Ng et al., Reference Ng, Loong, Liu and Weatherall2000; Lee and Hong-Kin, Reference Lee and Hong-Kin2005). If the mothers know their children's compliance behaviours do not reflect their true thoughts, it can result in negative feelings as well as combined ambivalence in mothers. Another possible reason is that if adult children always follow the advice of their mothers on important decisions, this may indicate an overall lack of independence in daily life. Prior studies suggest that parents feel more ambivalent about children when there is intergenerational dependence (Fingerman et al., Reference Fingerman, Chen, Hay, Cichy and Lefkowitz2006; Birditt et al., Reference Birditt, Fingerman and Zarit2010).
Limitations
A few limitations should be noted. First, because our sampling methods over-sampled women, our results might be limited in their generalisability. Second, focusing on ambivalent relations from mothers’ perspectives might lead to some bias in the results, because a previous study has found a larger gap in ambivalence between elderly mothers and children, and parents’ negative sentiments and ambivalence towards the relationship were lower compared to those of their children (Willson et al., Reference Willson, Shuey, Elder and Wickrama2006). Thus, further study could include ambivalence from both mothers’ and children's perspectives, which would more accurately depict the relationship between filial piety and intergenerational ambivalence. Third, this study covered most aspects of filial piety but lacked a measurement for respect, so that we used a reversed score for conflict frequency as a substitute. In addition, this study could not establish causality between filial piety and intergenerational ambivalence using cross-sectional data.
Conclusions
This study has shown that the ambivalence theory is applicable in explaining the complex intergenerational relationship in rural China, and characteristics of filial piety overlap with intergenerational solidarity as a source of ambivalence. In this study, some aspects of filial piety were associated with elderly mothers’ ambivalence, while most aspects of children's more filial behaviours and attitudes contributed to mothers’ positive feelings. Though the intergenerational relationship in rural China continues to change with urbanisation and population mobility, our study confirms that maintaining the tradition of filial piety can still benefit intergenerational relations. This study may add to our knowledge regarding the association between filial piety and ambivalence in rural China. Future studies might consider more research about the local context of culture.
Author ORCIDs
Qiuju Guo, 0000-0001-8868-4395.
Acknowledgements
The authors wish to thank all those who kindly volunteered to participate in the study.
Author contributions
The manuscript was written through contributions of all authors. QG designed the study, analysed the data and wrote each section in the paper. XG contributed to statistical analyses and the methods section. FS contributed to the introduction, and the results and discussion sections. NF contributed to the literature review section. All authors have given approval to the final version of the article.
Financial support
This work was supported by the National Social Science Foundation of China (15CRK003).
Conflict of interest
The authors declare no conflicts of interest.
Ethical standards
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. All the participants were voluntary and informed consent was obtained from all participants in the study.
