Youth Social Policy

International Sociology 28(3) 277 –296

© The Author(s) 2013 Reprints and permissions:

sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0268580913484345

iss.sagepub.com

Filial piety in contemporary Chinese societies: A comparative study of Taiwan, Hong Kong, and China

Kuang-Hui Yeh Academia Sinica, Taiwan

Chin-Chun Yi Academia Sinica, Taiwan

Wei-Chun Tsao Academia Sinica, Taiwan

Po-San Wan Chinese University of Hong Kong, Hong Kong

Abstract This study investigates the functions and implications of contemporary filial piety in three Chinese societies, namely, Taiwan, Hong Kong, and China, using large-scale cross-national datasets from the 2006 East Asian Social Survey. Despite the shared Confucian cultural values among these three societies, they have sharply differed in their paths toward modernization and in the development of their sociopolitical structures over the last century. The authors propose that the implications and influences of filial piety tend to be more similar in Taiwan and Hong Kong, but may be different in China because of profound differences in its sociopolitical system. Using the dual filial piety model as the baseline for comparative analyses, the results show that dual filial piety can be found in all three societies, although there are some componential alterations in China. The study also goes beyond the common practice of treating filial piety within the confines of caring for family elders by considering its functional utility to influence an individual’s sociopsychological outcomes. The regression results support the significance of dual filial piety and its close association with various aspects of daily life in contemporary Chinese societies.

Corresponding author: Kuang-Hui Yeh, Academia Sinica, Institute of Ethnology, Academia Sinica, Taipei, 115, Taiwan. Email: ykh01@gate.sinica.edu.tw

484345 ISS28310.1177/0268580913484345International SociologyYeh et al. 2013

Article

278 International Sociology 28(3)

Keywords Chinese culture, comparative study, dual filial piety model, East Asian Social Survey

Introduction

Filial piety has been a salient and productive research topic within Chinese societies. Despite having a similar culture rooted in Confucianism, the meaning and practice of contemporary filial piety in different Chinese societies may vary with the sociocultural context, which has sharply differed across societies in terms of the path toward modernization and in the devel- opment of sociopolitical structures over the last century. However, a comparative study on the importance and meanings of filial piety among different Chinese societies has yet to be conducted because of the difficulty in collecting equivalent data. The present study investi- gates the functions of contemporary filial piety in three Chinese societies, namely, Taiwan, Hong Kong, and China, using large-scale cross-national datasets from the 2006 East Asian Social Survey (EASS). In order to capture the diverse implications of filial piety in the three societies within a single theoretical framework, this article introduces a newly developed model called the dual filial piety model (DFPM; Yeh and Bedford, 2003) that conceptualizes filial piety by elucidating its underlying mechanisms and integrating a variety of inconsist- ent and conflicting findings from previous studies concerning the changing trends and con- sequences of filial piety. The present study empirically tests the extent to which the DFPM operates in each of the three societies to advance the contextualized understanding of filial piety in contemporary Chinese societies.

Dual dimensions of filial piety

Filial piety serves as the guideline for parent–child interaction and the traditional basis for harmonious intergenerational relationships within Chinese families. Considering the horizontal (equal relationship between two particular individuals) and vertical (hierar- chical relationship between different family roles) duality of parent–child relationships, the DFPM proposes two fundamental aspects of filial piety: reciprocal and authoritar- ian. These aspects represent two psychological motives that guide children’s interactions with their parents.

Reciprocal filial piety (RFP) is guided by spontaneous affection originating from long-term close interactions that fulfill the psychological need for mutual relatedness between two individuals (but not between the family role dyad). The basis of RFP is the intimacy and quality of the parent–child relationship. A positive relationship with parents fosters RFP, which entails gratitude and willingness to repay one’s parents’ care and sacrifice. From a historical perspective, RFP corresponds with ‘the natural inner disposi- tion of filial ethics’ emphasized during the pre-Chin era and with the Confucian principle of favoring the intimate (qin qin). RFP fits better with modern values such as democracy and equality and is much closer to the ‘pure relationship’ proposed by Giddens (1992) and thus is often expressed in voluntary support and expressions of love and care for one’s parents.

Authoritarian filial piety (AFP) is guided by obedience to normative authority (role of parents) and fulfills the need for collective identification. The parent–child relationship

Yeh et al. 279

is asymmetric in the beginning (Adams and Laursen, 2001) and parents will inevitably become ‘role models’ who represent ‘absolute authority’ during their children’s sociali- zation process. AFP is developed by children’s normative reaction to satisfy parental demands or expectations. However, parental authority often gradually weakens as chil- dren grow into adolescence, thus the performance of AFP requires additional reinforce- ment. Emphasis on AFP in Chinese history is closely related to the need to strengthen political sovereignty. AFP corresponds to the ideology of ‘patriarchal parental authority as a representation of the emperor’s absolute authority,’ which was legitimatized by the sovereigns from the Han to the Qing periods. AFP also corresponds to the Confucian principle of respecting the superior (zun zun). Hence, the performance of AFP often involves self-oppression and absolute obedience to authority.

According to previous findings, RFP is positively associated with females, a higher level of education, and with a higher socioeconomic status because of its connections with modern democratic values, while AFP is positively associated with males, older ages, lower educational background, and lower socioeconomic status because of its emphasis on patriarchal values (see Yeh, 2009a, for a review article on the develop- ment of the DFPM). Empirical evidence also shows that unlike RFP, AFP is positively correlated with traditional and conservative attitudes (e.g., male superiority and sub- mission to authority) (Yeh and Bedford, 2003). RFP, which is developed via intergen- erational communication and sharing in daily life, is often positively related to interpersonal skills (e.g., self-disclosure and empathy), better psychosocial adjust- ment, and emotional support to parents (Yeh, 2009a, 2009b). AFP is often positively related to maladaptation (e.g., neurotic personality traits, depression, and anxiety; Yeh, 2009a) and less related to emotional support of parents (Yeh, 2009b) because of its emphasis on self-oppression.

Another important finding from previous studies concerns the dual distinctive aspects of RFP and AFP, which are documented as not mutually exclusive but coexistent within an individual (Yeh, 2009b). These filial aspects provide the same function as a promoter of solidarity at the family level, although their underlying mechanisms are distinctive from each other at the individual level. For instance, both RFP and AFP tend to reduce parent–adolescent conflict at the family level, though the effect of RFP (via reconcilia- tion) is generally more significant than that of AFP (via inhibition) (see Yeh, 2009a). Hence, RFP and AFP may simultaneously function well to varying degrees contingent upon circumstances.

Although the dual dimensions proposed by the DFPM have been empirically sup- ported in Taiwan (see Yeh, 2009a), Hong Kong (Leung et al., 2010), and China (Jin et al., 2011), their relative importance and distinctive functions still need to be explored. The historical, social, and political contexts of each region are briefly introduced in the fol- lowing section as background for interpreting the subsequent statistical analyses.

A contextualized overview of the contemporary evolution of filial piety in Taiwan, Hong Kong, and China

This comparative overview of the sociocultural context in China, Taiwan, and, Hong Kong first reviews the local political system, modernization path, and some specific policies or

280 International Sociology 28(3)

historical events relevant to filial piety in each context, and then reviews the connection between contextual effects and dual filial piety.

China has been a totalitarian communist country emphasizing ultimate loyalty to the nation since 1949. The communist ideology that rejects parental authority and any other forms of social hierarchy such as gender stratification inevitably considers filial piety as authentically repulsive (Chow, 1991). Official statements of objection to filial values were issued during the Cultural Revolution from 1966 to 1976, mobilizing the young to rebel against their parents and severely harming the ethical basis for intergenerational solidarity. China is a latecomer in terms of modernization, having experienced rapid yet incomplete social transformation in the past 10 years due to dramatic economic growth subsequent to the ‘reform and opening up’ period. Changing filial values under the coun- try’s consolidated totalitarian regime are focused more on commercialization (Whyte, 1997) rather than democratization. However, the pace of modernization and its impact on filial piety are likely to be varied within China especially between urban and rural areas. Thus, the changing meaning of filial piety in China is ambiguous and lacks consensus. For instance, Whyte (1997) found an unexpected result of an unusually conservative filial attitude among young adults in the city of Baoding. In contrast, Zhan et al. (2008) proposed the concept broadened filial piety, which reinterprets filial norms to accom- modate emerging institutional care in Nanjing. Compared with Taiwan and Hong Kong, the distinguishing feature of contemporary filial piety in China may be more appropri- ately connected to the highly promoted policy, the Family Support Agreement (FSA), which emerged in rural areas in the mid-1980s. The FSA is signed by elderly parents and their children after negotiations and monitored by the government. It had been signed by more than 13 million rural families across China by the end of 2005, and is now finding its way into cities. As an official written intergenerational contract backed by legal sanc- tions and the principles of commercial exchange, the FSA has eroded spontaneity and affection in the practice of filial piety in China (Chou, 2011). Considering China’s mod- ernization process and the currently prevailing policy, the changing trend of filial piety in China may be better characterized as ‘commercialization of the intergenerational relationship.’

On the other side of the Taiwan Strait, the Kuomintang (KMT) retreated to Taiwan in 1949 after losing the civil war to the Chinese Communist Party. The KMT ruled Taiwan as a single-party state under martial law until 1990. Taiwan thus experienced a long and gradual process of political democratization and economic development instead of a radical disruption of traditional Chinese culture. Furthermore, to distinguish itself from Communist China, the Taiwanese government (KMT) regarded itself as the sole legiti- mate successor of traditional Chinese culture and initiated the Chinese Culture Renaissance Movement to preserve traditional Confucian values in modern society. This movement advocated for the advancement of filial piety along with the changing times by turning the focus on intergenerational affection instead of parental authority. According to a series of studies using multi-wave data from the Taiwan Social Change Survey, filial piety is still considered as the primary value in Taiwan (Huang and Chu, 2012; Jou and Chu, 2012) and the core meaning of ‘family harmony’ has shifted from ‘prior approval for collective welfare rather than individual interests’ (AFP) to ‘consid- eration of mutual affection and intimacy’ (RFP) (Yeh et al., 2012).

Yeh et al. 281

Hong Kong was a territory of China and a colony of Britain from 1842 to 1997. With its colonial background, the degree of westernization of Hong Kong is relatively high. Even though Hong Kong was returned to China in 1997, Hong Kong citizens still favor modern western values over communist ideology. Modern western values have perme- ated into all aspects of daily life because of the long-term British colonialism. Although the British colonial government did not advocate any particular policy to eliminate traditional Confucian values, a bicultural self-identity has been documented under the coexistence of socialization processes in both family (for the Chinese self) and school (for the western self) (Ng et al., 2007). Research indicates that Hong Kong Chinese often interpret their filial behaviors (e.g., caring for elderly parents) as a affection-based repayment toward parents (as the core of RFP) rather than obedience to the external norms (Wong and Chau, 2006). Most empirical evidence consistently shows that filial piety has gradually declined but never disappeared in Hong Kong (Lee and Kwok, 2005; Ng et al., 2002).

Previous research shows that RFP and AFP can be differentiated in all Chinese socie- ties. However, diversity in components may appear within each dimension because of interference from some coercive policies, especially in China. For example, one of the items measuring AFP – ‘One must have at least one son’ – originally reflected absolute obedience to traditional norms with reference to ‘Not having offspring is the most unfil- ial act’ (Mencius, one of the Confucian classics). However, its meaning may be reinter- preted as a violation of government policy because of the one-child policy in China. The regional heterogeneity within China due to urban–rural disparity in economic develop- ment and modernization may also have resulted in the lack of a stable and consistent form of filial piety across the nation. Hence, the components of RFP and AFP are sus- pected to demonstrate more variation in China than in Taiwan or Hong Kong. Filial piety has blended with modern western values in the latter two, but differentiation between the dual filial dimensions may reveal subtle differences between them. For example, in Taiwan, RFP has focused on intimate affection rather than equality, while AFP is still valued for its connection with affection-based family solidarity. In Hong Kong, RFP leans toward equal relationships rather than affection, while AFP, which reflects tradi- tional conservative values, has declined with westernization. The differentiation between RFP and AFP is expected to be more pronounced in Hong Kong.

The present study

This study aims to empirically verify the cross-societal validity of the DFPM in Taiwan, Hong Kong, and China with the 2006 family module of the EASS dataset. Items in the topical module of the EASS are selectively integrated into a pre-existing survey frame- work of each member country rather than conducted as an independent survey, limiting the number of common items and unintentionally disassembling some well-developed scales in the original module. However, large-scale data collected from adult samples are rare and so valuable for comparative research on the forms and functions of contempo- rary filial piety in Chinese societies.

We first conducted an exploratory factor analysis by region to investigate the cross- societal validity of the dual-dimension structure (RFP/AFP) as proposed by the DFPM.

282 International Sociology 28(3)

Next, we conducted multiple regression analyses by region to verify whether the influ- ence of the extracted filial factors on different types of criterion variables corresponded to the theoretical propositions of the DFPM. More specifically, both RFP and AFP may have a positive influence on the criterion variables regarding family elder care (attitudes toward adult children’s family obligations and actual intergenerational support behav- iors for parents), though RFP may have greater influence. RFP and AFP may have dif- ferent and even opposite influences on the criterion variables regarding individual aspects (RFP has a positive influence on personal adaptation and AFP a negative influ- ence; RFP has little or no negative influence on patriarchal attitude and traditional gen- der-division attitude and AFP has a positive influence). The descriptive statistics, the correlations with demographic variables, and the results of a one-way ANOVA (analysis of variance) for each filial item by region were used as additional information to under- stand and compare the relative importance and properties of each extracted filial factor in the three samples.

Method

Data and sample

Data were taken from the 2006 family module of the EASS. Corresponding datasets from China, Hong Kong, and Taiwan were collected through face-to-face interviews with a regional representative adult sample randomly chosen through a multistage strati- fied sampling method. More information about the sampling process can be found in the Study Description Form provided by each member country of EASS.1 Detailed informa- tion for the China dataset, which was based on the 2006 China General Social Survey, can be obtained online.2 Considering the difference in age range among the sampling frames of the three societies (17–69 years in China, 19–92 in Taiwan, and 18–88 in Hong Kong), only adult respondents between 20 and 69 years old (the overlapping age range among all three samples) were included in the final sample. Due to the increasing num- ber of internal migrants from China after the 1997 handover, 38 Hong Kong respondents who were not born or had lived fewer than five years in Hong Kong were deleted. To enhance the representativeness of the sample, the results presented in this study are based on the weighted data provided by the EASS dataset. The final weighted samples con- sisted of 3109 (48% males and 52% females) respondents from China, 821 (47.4% males and 52.6% females) from Hong Kong, and 1849 (49.7% males and 50.3 females) from Taiwan.

Measures

Target variable: dual filial piety beliefs. The dual filial beliefs were measured with items from the short-form dual filial piety scale, which has been empirically supported as a valid measure to investigate filial beliefs across cultures (Yeh, 2009a; Yeh and Bedford, 2003). Although the original scale has nine items, only six were used in this study due to limitations with the EASS dataset (see Table 1, the first three items originally indicated RFP, and the rest AFP). For RFP, three items measured authentic gratitude, caring, and

Yeh et al. 283

disinterested sacrifice for parents based on perceived parental love during childhood. For AFP, three items measured the absolute priority of parental expectations and family wel- fare. Respondents reported their degree of agreement on a seven-point scale ranging from 1 (strongly disagree) to 7 (strongly agree). A higher score denoted a higher level of filial belief.

Demographic control variables relevant to dual filial beliefs Individual demographic variables

1. Gender was coded as ‘1’ for male and ‘0’ for female. 2. Age was a continuous variable ranging from 20 to 69 years. 3. Educational level was measured by ‘the highest education level’ with six catego-

ries (1 = no formal qualification, 2 = lowest formal qualification, 3 = above low- est qualification, 4 = higher secondary completed, 5 = above higher secondary level, 6 = university degree completed).

4. Personal income: the region-specific variable of the respondent’s average monthly income was chosen as the income indicator for each of the samples. To increase comparability, income data from each sample were converted to z-scores.

5. Social class: respondents from China and Hong Kong evaluated their own social class on a five-point scale from 1 (lower) to 5 (upper), whereas Taiwanese

Table 1. Descriptive statistics of each filial item by region and cross-societal differences in each filial item.

Filial piety items Taiwan Hong Kong

China ANOVA Scheffé post- hoc

Mean SD Mean SD Mean SD F(2, 5775)

1. Be grateful to parents for raising me

6.66 0.62 6.23 0.75 5.99 0.94 386.77*** TW>HK>CN

2. No matter what my parents do, I’ll still treat them well

6.10 1.12 5.79 1.04 5.81 0.94 53.33*** TW>HK & CN

3. Support parents to make them more comfortable

6.29 0.84 5.84 0.88 4.64 1.35 1295.44*** TW>HK>CN

4. To continue the family line, one must have at least one son

4.24 1.83 3.65 1.65 4.54 1.48 99.83*** CN>TW>HK

5. One should give up personal interests to fulfill parental expectations

3.85 1.65 3.44 1.48 5.77 0.94 1831.87*** CN>TW>HK

6. Children must do something to honor their parents

5.79 1.23 4.98 1.33 5.63 0.99 150.04*** TW>CN>HK

*p < .05, **p < .01, ***p < .001.

284 International Sociology 28(3)

respondents used a 10-point scale ranging from 1 (lowest) to 10 (highest). To attain comparability, region-specific social class variables from each sample were converted into a z-score.

6. Occupational status: a score of 1 to 5 was assigned. Higher scores represented higher occupational status. The five categories were: (1) elementary occupation or unskilled worker, (2) plant or machine operator, craft worker, and skilled agri- cultural or fishery worker, (3) personal service, sales, or office clerk, (4) profes- sional, technician, or associate professional, and (5) legislator or senior official, manager, or chief.

Family-demographic variables

1. Household size was measured as a continuous variable based on the reported number of co-resident family members.

2. Marital status was recoded as ‘0’ for single (including never married, divorced, or separated) or ‘1’ for married (including married, widowed, or cohabitating).

3. Number of parents alive: a score of 1 to 3 was given to ‘Both parents deceased,’ ‘One deceased and one living’, or ‘Both parents still alive.’ A higher score denoted higher potential demand for support from parents.

Criterion variables relevant to dual filial beliefs Attitude toward traditional gender division in the family. This variable was composed of

three items: ‘A husband’s job is to earn money; a wife’s job is to look after the home and family’; ‘If husband’s family and wife’s family need help at the same time, a married woman should help the husband’s family first’; and ‘It is more important for a wife to help her husband’s career than to pursue her own career.’ Respondents indicated their agreement with each item on a seven-point scale ranging from 1 (strongly disagree) to 7 (strongly agree). A higher score represented a stronger attitude toward traditional gender division, which may be a basis for harmonious family functioning.

Patriarchal attitude. This variable indicated male superiority within the domains of family and work and consisted of two items: ‘The eldest son should inherit a larger share of the property’ and ‘During economic recession, it is all right for women to be laid-off prior to men.’ Respondents indicated their agreement with each item on a seven-point scale ranging from 1 (strongly disagree) to 7 (strongly agree). A higher score represented a stronger patriarchal attitude.

Personal adaptation. Personal adaptation was composed of two items using a five- point scale: self-rated health (ranging from 1 [very bad] to 5 [very good]), and overall life satisfaction (ranging from 1 [very dissatisfied] to 5 [very satisfied]). A higher score represented better physical, psychological, and social functioning in daily life.

Attitudes toward adult children’s family obligations. Two items measured attitude toward unmarried and married adult children’s obligations to provide financial support for their own parents and a second pair measured attitude toward the obligations of married men and married women to their spouse’s parents. Respondents reported their agreement with all four items on a seven-point scale ranging from 1 (strongly disagree) to 7 (strongly agree). A higher score denoted greater endorsement of adult children’s obligation to their parents or spouse’s parents.

Yeh et al. 285

Actual intergenerational support behaviors toward parents. This variable comprised three items regarding the different kinds of intergenerational support actually provided by adult children in the past year: financial support, household chores, and emotional support. Respondents (except those whose parents had passed away) reported their actual behavioral frequency for each item on a five-point scale ranging from 1 (not at all) to 5 (very frequently). A higher score represented more intergenerational support.

Results and discussion

Filial dimension comparison based on exploratory factor analysis results

Principal axis factor analysis of the six filial piety items was conducted by region. The criterion to determine the appropriate number of factors was set according to the ‘eigen- values-greater-than-1’ rule. As a medium positive correlation between RFP and AFP was found in previous studies (see Yeh, 2009a), a promax method for oblique rotation was adopted.

The original rotated two-factor solution approximately supported the DFPM in both Taiwan and Hong Kong. The two extracted factors corresponded to the RFP and AFP dimensions, which in turn accounted for 58.5% and 55.3% of the total variance in Taiwan and Hong Kong, respectively. The result for the Hong Kong sample (with pri- mary loadings ranging from .42 to .78) was congruent with the dual dimensions pro- posed by the DFPM. A cross-loading problem on item 6, which theoretically pertained to AFP but had equal loadings (.47) on both factors, was found in the Taiwan sample. However, the other five items all had primary loadings over .56 without cross- loadings. Overall, the factor structure extracted from the Taiwan data did not perfectly match the dual dimensions proposed by the DFPM. However, the factor structure was still acceptable. Whether the unexpected cross loading for item 6 in the Taiwan sample was a consequence of recent social change remains to be investigated. The RFP and AFP Cronbach’s alphas were .64 and .54 respectively in the Taiwan sample, and .63 and .52 in the Hong Kong sample.

China’s original rotated two-factor solution, which did not correspond to the DFPM dimensions, accounted for 63.4% of the total variance. The first extracted fac- tor (labeled as FP1 and the likely counterpart of RFP) consisted of four items (items 1, 2, 5, and 6) with primary loadings above .57; the second extracted factor (labeled as FP2 and the likely counterpart of AFP) had a loading of .56 for item 4 and .50 for item 3. The FP1 and FP2 Cronbach’s alphas were .79 and .48, respectively. The seem- ingly low FP2 alpha may be due to its limited number of items (only two). Although both extracted factors were comprised of RFP and AFP items, their statistical charac- teristics still reflected differentiation between AFP and RFP. The data from Taiwan and Hong Kong (Table 1) matched previous findings where each RFP item had a higher mean and a lower standard deviation (SD) than each AFP item (Leung et al., 2010; Yeh, 2009a). FP1 items in China corresponding to RFP (items 1 and 2) and AFP (items 5 and 6) maintained the expected pattern of score mean and SD. We therefore argue that the variation in factor structure is caused by the unusually low scores of the two items in FP2.

286 International Sociology 28(3)

According to family-relevant policies in contemporary China, the low scores on the FP2 items (‘have at least one son for family lineage’ and ‘support parents’ livelihood to make them more comfortable’) may be an outcome of the one-child policy and the FSA mentioned previously. The one-child policy not only directly blocked the possibility of having more than one child, but also weakened the intention of children to provide long- term support for their parents (Tu et al., 1989). As the FSA has been promoted by the Chinese government as a legally enforceable contract, the result was a passive obedience rather than increased voluntary provision of support for parents (Xu, 2001). Thus, the FP2 may have emerged from the coercive impact of official policies on personal filial beliefs.

Despite alterations in their composing items, we argue that the two filial factors from the China data still show a difference between voluntary behaviors toward parents and normative intergenerational relationships regulated by legal coercion, that corresponds to the dual dimensions of the DFPM. To verify the correspondence between FP1 and RFP and between FP2 and AFP, subsequent analyses of the China data were conducted with the originally extracted factor structures – FP1 and FP2.

One-way ANOVA results

A one-way ANOVA was used to test for differences in the mean of each filial item among the three Chinese societies. Results are shown in Table 1. Scheffé post-hoc comparisons for each item indicated that items 1 to 3 showed similar patterns of cross-regional differ- ences. The Taiwan sample gave a significantly higher importance rating for these filial items than the Hong Kong sample who gave a significantly higher rating than the China sample. Item 2, on which China and Hong Kong showed no significant difference, was the only exception. Items 4 to 6 also showed a similar pattern of cross-regional differ- ences. The China sample gave a significantly higher rating for these filial items than the Taiwan sample, which gave a significantly higher rating than the Hong Kong sample. The only exception was item 6, in which the Taiwan sample gave a significantly higher rating than the China sample.

These item score patterns reflect a particular correspondence with contextual effects. As a consequence of the Chinese Culture Renaissance Movement, average scores on RFP items (items 1 to 3) were relatively high in Taiwan. The unexpected high score on item 6 in Taiwan may be a result of the gradual shift in the meaning of gaining social achievements from ‘honoring one’s family and clan’ to ‘repayment for parents due to gratitude.’ In Hong Kong, the average scores on AFP items (items 4 to 6) were relatively low because of the high level of westernization. The unexpected low score on RFP item 2 of the Hong Kong sample may be the result of the peculiar emphasis on equality. Thus, item 2 may be interpreted as an undemocratic request for the sacrifice of children, rather than for unconditional love and dedication. However, the RFP of the Taiwan and Hong Kong samples showed greater importance than AFP, similar to the findings of previous studies (see Yeh, 2009a; Yeh et al., 2012).

Notably, the average scores of all filial items in the China sample were relatively close to each other. The differentiation between FP1 and FP2 seemed vaguer than that between RFP and AFP. However, the average score of almost all filial items in all three regions

Yeh et al. 287

(except very few items in the Hong Kong and Taiwan samples) was above the midpoint of 4 on the seven-point scale. Filial piety therefore remains widely valued in contempo- rary Chinese societies.

Correlation and multiple regression analyses results

Correlation analysis. Table 2 presents the correlation between the demographic variables that served as controls in the regression analyses and the filial factors extracted from each of the three samples. In Taiwan and Hong Kong, most of the demographic variables had higher correlations with AFP than RFP. That is, AFP was significantly related to older age, male gender, low education and social status, larger household sizes, and to the married. In the China data, FP2 (the counterpart of AFP) had higher correlations with the demographic variables than did FP1 (the counterpart of RFP). The relationship between FP1 and FP2 showed a significantly low-to-moderate positive correlation (r = .19, p < .001), corresponding to the correlation between RFP and AFP extracted both from the Taiwan (r = .34, p < .001) and Hong Kong (r = .20, p < .001) samples. The results supported the non-mutually exclusive relationship between RFP and AFP, and also the similarities between the dual-factor structure in the China sample (FP1/FP2) and the REP/AFP framework proposed by the DFPM.

Differences between RFP and AFP in Taiwan and Hong Kong in correlation patterns with demographic variables manifested some anticipated contextual effects. The differ- ence between RFP and AFP was sharper in Hong Kong than in Taiwan. That RFP in Hong Kong did not correlate with each demographic variable may reflect the fact that RFP has been widely accepted by people with different social characteristics because of the high level of modernization and westernization. In contrast, the differentiation

Table 2. Correlations between demographic variables and filial factors extracted from each sample.

Taiwan Hong Kong China

RFP AFP RFP AFP FP1 FP2

Age .13*** .22*** −.00 .15*** −.04* .131*** Gender −.00 .12*** −.04 .11** .01 .02 Education level −.08*** −.29*** .02 −.22*** .08*** −.20*** Social class identification

.08*** −.02 .02 −.09* .02 −.07***

Personal income .02 .01 .07 −.06 .02 −.09*** Occupational status −.01 −.18*** .05 −.11* .04* −.12*** No. of parents alive −.07** −.21*** −.05 −.11** .03 −.08*** Household size .06* .10*** .02 .10*** .02 .10*** Marital status .12*** .14*** .03 .10** −.02 .10*** Correlations between filial factors

.34*** .20*** .19***

*p < .05, **p < .01, ***p < .001.

288 International Sociology 28(3)

between RFP and AFP in Taiwan was indefinite because of the Chinese Culture Renaissance Movement. Although in Taiwan RFP had similar correlation patterns as AFP with some demographic variables, most of these results could be attributed to the DFPM. The only unexpected result was the low but significant correlation between RFP and educational level (r = −.08, p < .001) in the Taiwan sample. Further investigation is required to ascertain the cause.

Multiple regression analysis. The corresponding features between the filial factor struc- tures of RFP/AFP and FP1/FP2 were investigated through multiple regression analyses by region. This method was applied to examine their relative effects (both direction and strength) on the selected criterion variables while controlling for major sociodemo- graphic variables. All the regression models were significant. Results are shown in Tables 3 and 4.

In the Taiwan and Hong Kong samples, the impact of RFP and AFP on the criterion variables was in accordance with the DFPM, despite a few exceptions. The impact of FP1 and FP2 in the China sample also corresponded with RFP and AFP, respectively. First, AFP and FP2 had a significant positive influence on attitudes toward gender divi- sion and patriarchy (with βs ranging from .35 to .46). In contrast, RFP and FP1 had no significant influence on gender-division attitude, and had significant negative influence on patriarchal attitude (with βs ranging from .15 to .17). The two criterion variables regarded as traditional and socially conservative attitudes had stronger connections with AFP than RFP, as found previously (e.g., Yeh and Bedford, 2003). Associations of RFP (or FP1) with gender-division attitude differed from patriarchal attitude because the for- mer defines the foundation of family functions and mutual support between couples in traditional Chinese culture. Hence, RFP, which focuses on equal relationships and pro- motes family solidarity at the collective level, may be less opposed to gender role divi- sion than to patriarchy. This finding also explains the low positive impact of FP1 (β = .05, p < .05), the RFP counterpart, on gender-division attitude in the China sample.

Second, RFP and FP1 showed greater effects on personal life adaptation than AFP and FP2 despite the non-significant effects of both filial factors in the Hong Kong sample. Although personal adaptation was strongly associated with young, high-class respondents in all three societies, RFP and FP1 still had a significant positive influence in Taiwan (β = .10, p < .00) and China (β = .06, p < .01). In contrast, AFP had less positive influence (β = .07, p < .05) on personal adaption in Taiwan, and FP2 had a marginally significant negative influ- ence (β = −.03, p < .1) in China. The difference between RFP (or FP1) and AFP (or FP2) in these two samples accorded with the DFPM. In Hong Kong, the non-significant results of the dual filial factors might have been caused by the westernization of its social context. The function of filial piety may be limited to a few aspects of intergenerational interaction.

Finally, RFP (or FP1) showed more significant positive influence on elder care, including both relevant attitudes and actual support behaviors across the three samples as compared with AFP (or FP2). The slightly stronger influence of AFP on attitudes toward providing financial support to a spouse’s parents in Hong Kong was the exception. At the attitude level, after controlling for sociodemographic variables, both RFP and FP1 had greater positive influence on attitudes toward one’s obligation to provide financial sup- port to one’s parents and one’s spouse’s parents. AFP and FP2 had no effect or lower

Yeh et al. 289

T a b

le 3

. M

ul ti pl

e re

gr es

si o n

o f pe

rs o na

l v al

ue s

an d

ad ap

ta ti o n

o n

fil ia

l p ie

ty c

o nt

ro lli

ng f o r

de m

o gr

ap hi

c fa

ct o rs

.

Pe rs

o na

l v al

ue s

Pe rs

o na

l a da

pt at

io n

G

en de

r- di

vi si

o n

at ti tu

de Pa

tr ia

rc ha

l a tt

it ud

e

T

W H

K C

N T

W H

K C

N T

W H

K C

N

A ge

.0 6+

− .0

3 .0

0 .0

6 −

.0 6

.0 5+

− .1

6* **

− .1

1 −

.1 8*

** G

en de

r .0

5* .0

9+ .0

8* **

.1 3*

** .1

0* .1

0* **

− .0

1 −

.0 6

.0 3

Ed uc

at io

n le

ve l

− .2

5* **

− .0

8 −

.0 8*

** −

.0 2

.0 8

− .0

3 .0

2 .0

5 .0

5* So

ci al

c la

ss

id en

ti fic

at io

n .0

1 .0

2 −

.0 2

.0 3

.0 4

.0 0

.1 7*

** .2

0* **

.1 5*

**

Pe rs

o na

l i nc

o m

e −

.0 2

− .0

8 −

.0 2

− .0

9* *

− .0

3 .0

1 .0

7* .1

2* .0

5* O

cc up

at io

na l s

ta tu

s −

.0 7*

− .1

3* −

.0 2

− .0

5 −

.1 9*

** .0

1 .0

4 .0

4 .0

3 N

o . o

f pa

re nt

s al

iv e

− .0

5 −

.0 4

.0 7*

* .0

4 .0

2 .0

4 −

.0 2

− .0

1 .0

6* H

o us

eh o ld

s iz

e −

.0 2

.0 6

.0 1

.0 1

− .0

2 −

.1 0*

** .0

0 −

.0 7

− .0

2 M

ar it al

s ta

tu s

.0 8*

* .1

4* .0

3 −

.0 2

.1 0+

.0 0

.1 0*

* .0

2 .0

7* *

R F

P /F

P 1

.0 2

.0 7

.0 5 *

− .1

7 **

* −

.1 7 **

* −

.1 5 **

* .1

0 **

* .0

5 .0

6 **

A F

P /F

P 2

.3 7 **

* .3

5 **

* .3

6 **

* .3

5 **

* .4

6 **

* .4

5 **

* .0

7 *

.0 1

− .0

3 +

R 2

.3 8*

** .2

7* **

.1 7*

** .1

8* **

.2 8*

** .2

4* **

.0 9*

** .1

1* **

.1 2*

**

+ p

< .1

, * p

< .0

5, *

*p <

.0 1,

* **

p <

.0 01

.

290 International Sociology 28(3)

T a b

le 4

. M

ul ti pl

e re

gr es

si o n

o f cr

it er

ia r

eg ar

di ng

c ar

in g

fo r

fa m

ily e

ld er

ly o

n fil

ia l p

ie ty

c o nt

ro lli

ng f o r

de m

o gr

ap hi

c fa

ct o rs

.

A tt

it ud

es t

o w

ar d

ad ul

t ch

ild re

n’ s

o bl

ig at

io n

o f pr

o vi

di ng

f in

an ci

al s

up po

rt t

o

pa re

nt s

A ct

ua l i

nt er

ge ne

ra ti o na

l s up

po rt

be

ha vi

o rs

t o w

ar d

pa re

nt s

T

o o

ne ’s

o w

n pa

re nt

s T

o s

po us

e’ s

pa re

nt s

T

W H

K C

N T

W H

K C

N T

W H

K C

N

A ge

− .1

0* −

.1 2+

− .0

7* .0

0 .0

8 −

.0 3

.0 1

− .0

4 .0

3 G

en de

r −

.0 8*

* −

.1 8*

* −

.0 0

− .0

5 −

.1 7*

** −

.0 1

− .0

1 −

.1 4*

.1 3*

** Ed

uc at

io n

le ve

l .0

9* .0

0 .0

5* .0

5 −

.0 3

.0 9*

** .1

6* **

.0 8

.1 1*

** So

ci al

c la

ss

id en

ti fic

at io

n −

.0 0

− .0

9+ .0

3 .0

4 −

.0 5

− .0

2 .0

5+ −

.0 2

.0 9*

**

Pe rs

o na

l i nc

o m

e −

.0 3

.0 7

− .0

3 −

.0 5

.0 9

− .0

3 .0

9* *

.1 3+

.0 3

O cc

up at

io na

l s ta

tu s

.0 3

.0 2

.0 2

.0 5

.0 1

.0 1

− .0

2 .0

5 −

.0 7*

N o . o

f pa

re nt

s al

iv e

.0 2

− .0

7 −

.0 2

.0 1

− .0

5 .0

3 −

.0 5

.0 9

− .0

5+

H o us

eh o ld

s iz

e −

.0 0

− .1

4* *

.0 3+

− .0

2 .0

3 .0

2 .0

6* .0

5 .0

7* *

M ar

it al

s ta

tu s

− .0

6+ .0

1 −

.0 8*

** −

.0 3

− .0

0 −

.0 6*

* −

.0 6

− .1

0 −

.0 7*

* R

F P

/F P

1 .2

4 **

* .1

4 **

.2 5 **

* .1

7 **

* .1

0 +

.2 3 **

* .1

9 **

* .1

6 **

.1 1 **

* A

F P

/F P

2 .0

0 .0

5 .0

1 .0

3 .1

1 *

.0 6 **

.0 7 *

− .0

1 .0

3 R

2 .1

0* **

.1 0*

** .0

8* **

.0 4*

** .0

7* *

.0 8*

** .0

9* **

.1 3*

** .0

7* **

+ p

< .1

, * p

< .0

5, *

*p <

.0 1,

* **

p <

.0 01

.

Yeh et al. 291

positive influence. These results also highlighted some subtle cross-societal differences. In Taiwan and Hong Kong, RFP had a relatively stronger influence on attitudes regarding obligation toward one’s own parents (with βs .24 and .14, respectively) than toward a spouse’s parents (with βs .17 and .10, respectively). These results reflected the affective core of RFP because the relationship with one’s own parents is more familiar than with one’s spouse’s parents. In contrast, the influence of FP1 in China on attitudes regarding obligation toward one’s own parents and one’s spouse’s parents were similar in size. The one-child policy seems to have made individuals more aware of the similar role and responsibility of (future) spouses, who might also be a single child, raising the impor- tance of filial obligation toward parents-in-law, as a previous study (Deutsch, 2006) indi- cated. Further, the difference in the influence of AFP between Taiwan (β = .03, p > .1) and Hong Kong (β = .11, p < .05) on attitudes regarding obligation toward parents-in-law showed some contextual differences as well. Under the patriarchal system in traditional Chinese culture, only married women have filial responsibility to their parents-in-law by social norm. The Chinese Culture Renaissance Movement in Taiwan, which advocated transforming the foundation of filial piety into affection, may have strengthened the effect of RFP instead of AFP on obligation toward parents-in-law for both genders. Since the core of RFP in westernized Hong Kong focuses on equality, filial obligations toward parents-in-law without any reciprocity or previous interaction may be regarded as unrea- sonable traditional norms for both genders rather than as expressions of affection, thus they are more associated with AFP.

At the behavioral level, RFP and FP1 had significant positive influence (with βs ranging from .11 to .19) on actual support behaviors toward parents across the three samples. In contrast, AFP and FP2 had no significant influence, except the low but significant positive influence of AFP (β = .07, p < .05) in the Taiwan sample. These results are in accordance with a previous finding that RFP has broader and greater effects on various support behav- iors toward parents than AFP (see Yeh, 2009b). The positive significant influence of both RFP and AFP in Taiwan showed similar functions of elder care, whereas the difference in their effect size supported the higher relative importance of RFP. The original RFP item regarding the maintenance of parents (item 3) was not subordinated to FP1 (as the counter- part of RFP) in China. Thus, the significant positive association between FP1 and actual support behaviors toward parents was the result of the underlying functioning mechanism of FP1, rather than the similarly superficial contents between filial items and criterion measures. Clearly, the dual fundamental dimensions of filial piety (RFP/AFP) and their functioning are consistent across societies, even though the meaning of a specific item and the components of each filial dimension are somewhat affected by contextual factors.

In Taiwan and Hong Kong, the core function of AFP represents the observance of conservative traditional norms, thereby strengthening patriarchal attitudes. In contrast, RFP represents affection-based gratitude from long-term intergenerational interaction with intimacy, thereby increasing attitudes and behavior supporting elderly parents. Correspondingly, FP1, which was more positively associated with criteria regarding elder care and personal adaptation, reflected the core characteristics of RFP, whereas FP2, which was more associated with patriarchal attitudes, reflected the core character- istics of AFP. Despite the difference in underlying items between RFP and FP1 and between AFP and FP2, the mechanisms of the dual filial factors extracted from the China sample were consistent with RFP and AFP.

292 International Sociology 28(3)

Conclusion

General discussion

This study examined contemporary filial piety to reveal its diverse evolutionary paths in different Chinese societies. Using the DFPM as the baseline for comparative analyses of China, Taiwan, and Hong Kong, our analyses showed that filial piety, especially RFP (or its counterpart, FP1), remains potent among all three societies despite the diverse socio- political developments in each society. Like previous findings in Taiwan (see Yeh, 2009a; Yeh et al., 2012) and Hong Kong (Ng et al., 2002; Wong and Chau, 2006), our results confirm that filial piety has not been eroded by modernization and democratization. Its core essence in Taiwan and Hong Kong has shifted from absolute submission and paren- tal authority to mutual affection and equality in the parent–child relationship. Some doubted the survival of filial piety under Chinese communism (Chow, 1991; Whyte, 1997). Our study suggests that even after the Socialist Transformation Movement and the radical Cultural Revolution, filial piety remains important in China. Incorporating various sociopsychological criterion variables at the individual level, we are able to cap- ture the implications of filial piety beyond the common practice of treating filial piety within the confines of family elder care. The fact that in all three samples each filial fac- tor had the highest impact on the theoretically corresponding criterion in the regression models also supports the significance of dual filial piety and its close association with daily life in contemporary Chinese societies.

The filial factor structure of China shows some componential alteration from the DFPM, but the distinctive functioning characteristics of FP1 and FP2 shown in both cor- relation and regression results corresponded to that of RFP and AFP, respectively. Hence, the altered dual factor structure in China still manifests the important differential func- tioning of affection-based gratitude and submission to absolute authority. Since some studies have provided empirical evidence for the dual structure of RFP and AFP with student samples (from middle school to college) in China (e.g., Jin et al., 2011; Li and Xu, 2011), the somewhat altered factor structure in China may just reflect the temporary instability in the changing trends of filial piety because of rapid modernization and con- flicting family policies. Filial piety may evolve toward a differentiation between RFP and AFP in succeeding generations.

This study shows that the contemporary filial piety in Chinese societies can be dif- ferentiated into at least two distinguishable aspects with distinctive functions, as pro- posed by the DFPM. Filial piety not only corresponds to salient psychological implications at the individual level in addition to its cultural impact, but also reflects the influence of societal and political contextual factors.

Limitations and future research

There are two major limitations of this study. First, the smaller number of common items across the datasets from the three Chinese societies somewhat weakens the connection between the theoretical model and the dataset in use. The limited number of items restricts the construction of research variables, especially dual filial piety (missing three items

Yeh et al. 293

from the original scale) and personal adaptation (missing multidimensional components). The result may have been an altered factor structure in China, the cross-loading of item 6 in Taiwan, and some unexpected regression results for personal adaptation. The limited number of items (between two and three) for each variable may also have led to the gener- ally low internal consistency reliability of each measure. Second, the intra-sample hetero- geneity in the China data may have caused difficulty in getting representative results for the whole nation. The excluded samples due to age filtering and missing values (espe- cially the items regarding actual intergenerational support) also reduced the size and rep- resentativeness of the final sample of each society, thus affecting the generalizability of the results. Furthermore, face-to-face interviews are more likely to be affected by social desirability bias than any other survey method.

Despite these limitations, a significant strength of the present study is the use of a comparative cross-societal framework, which can provide a better understanding of the functioning of filial piety in different Chinese societies. For instance, without compari- son to other samples, the lowest mean score of item 4 among all items in China could be interpreted as the declining belief toward continuing the family line. However, this item’s mean score in China was actually higher than in Taiwan or Hong Kong. Using the DFPM as the theoretical basis for interpretation of the comparative results allowed for more detailed exploration of the contemporary evolution of filial piety in each society, such as the subtle difference in the relative importance of RFP and AFP between Taiwan and Hong Kong, which is caused by the difference in degree of westernization.

In sum, our findings on filial piety and its functioning in the three societies echo a previous proposal that stated that cross-national similarities and differences may be simultaneously linked with multiple sources including individual agency (micro-level), family climates (meso-level), and relevant policies or social structure (macro-level), and should be cautiously interpreted (Lowenstein and Daatland, 2006). As filial piety contin- ues evolving with social trends, empirical investigation on filial piety in a comparative cross-societal context should be continued to attain a more integrated understanding.

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Notes

1. www.eassda.org/modules/doc/index.php?doc=non_01_01&___M_ID=93 2. www.uchicago.cn/wp-content/uploads/2011/05/Weidong-Wang.pdf

References

Adam R and Laursen B (2001) The organization and dynamics of adolescent conflict with parents and friends. Journal of Marriage and Family 63(1): 97–110.

Chow N (1991) Does filial piety exist under Chinese communism? Journal of Ageing and Social Policy 3(1–2): 209–220.

Chou RJA (2011) Filial piety by contract? The emergence, implementation, and implications of the ‘Family Support Agreement’ in China. The Gerontologist 51(1): 3–16.

294 International Sociology 28(3)

Deutsch FM (2006) Filial piety, patrilineality, and China’s one-child policy. Journal of Family Issues 27(3): 366–389.

Giddens A (1992) The Transformation of Intimacy: Sexuality, Love and Eroticism in Modern Societies. Cambridge: Polity.

Huang LL and Chu RL (2012) Is it the wind shearing or the tide going in and out? Searching for core values and its changing trend in Taiwan. The Journal of Kaohsiung Behavior Sciences 3: 60–90.

Jin CC, Zou H and Yu YB (2011) The trait of filial piety belief and the relationship of filial piety belief, attachment and interpersonal adaptation of middle school students. Psychological Development and Education 6: 619–624.

Jou YH and Chu RL (2012) The social change and determinant of the values of Taiwan people. In: Fourth International Conference on Sinology, Taipei, Taiwan, 20–22 June 2012.

Lee WKM and Kwok HK (2005) Differences in expectations and patterns of informal support for older persons in Hong Kong: Modification to filial piety. Ageing International 30(2): 188–206.

Leung AN-M, Wong SS-F, Wong IW-Y and McBride-Chang C (2010) Filial piety and psycho- logical adjustment in Hong Kong Chinese early adolescents. Journal of Early Adolescence 30(5): 651–667.

Li Q and Xu H (2011) The relationship among Chinese college students’ psychological traditional- ity and psychological modernity, trait empathy, authoritarian personality and dual filial piety. Psychological Exploration 31(6): 539–543.

Lowenstein A and Daatland SO (2006) Filial norms and family support in a comparative cross- national context: Evidence from the OASIS study. Ageing and Society 26(2): 203–223.

Ng CY, Philips DR and Lee WK (2002) Persistence and challenges to filial piety and informal support of older persons in a modern Chinese society: A case study in Tuen Mun, Hong Kong. Journal of Aging Studies 16(2): 135–153.

Ng SH, Yam N and Lai J (2007) The bicultural self of Chinese in Hong Kong. In: Liu J, Ward C, Bernardo A, et al. (eds) Casting the Individual in Societal and Cultural Contexts: Social and Societal Psychology for Asia and the Pacific. Seoul, Korea: Kyoyook-Kwahak-Sa, pp. 105–122.

Tu EJC, Liang J and Li S (1989) Mortality decline and Chinese family structure: Implications for old age support. Journal of Gerontology: Social Science 44(4): 157–168.

Whyte MK (1997) The fate of filial obligations in urban China. The China Journal 38(1): 1–31. Wong O and Chau B (2006) The evolving role of filial piety in eldercare in Hong Kong. Asian

Journal of Social Science 34(4): 600–617. Xu YB (2001) Family support for older people in rural China. Social Policy and Administration

35(3): 307–320. Yeh KH (2009a) The dual filial piety model in Chinese culture: Retrospect and prospects.

Indigenous Psychological Research in Chinese Societies 32: 101–148. Yeh KH (2009b) Revisiting some key issues of the dual filial piety model. Indigenous Psychological

Research in Chinese Societies 32: 207–248. Yeh KH and Bedford O (2003) A test of the dual filial piety model. Asian Journal of Social

Psychology 6(3): 215–228. Yeh KH, Chang YH and Tsao WC (2012) The change of family value and its underlying psycho-

logical mechanism in contemporary Taiwan. In: Yi CC and Chang YH (eds) Social Change in Taiwan, 1985–2005, Volume I: On Family, Taiwan Social Change Survey Symposium Series. Taipei: Institute of Sociology, Academia Sinica, pp. 59–101.

Zhan HJ, Feng X and Luo B (2008) Placing elderly parents in institutions in urban China. Research on Aging 30(5): 543–571.

Yeh et al. 295

Author biographies

Kuang-Hui Yeh is a research fellow and deputy director at the Institute of Ethnology, Academia Sinica, Taiwan. He is also a joint-appointment professor of the Department of Psychology, National Taiwan University. He was the editor-in-chief of Formosa Journal of Mental Health (2008–2010).

Chin-Chun Yi is a research fellow at the Institute of Sociology, Academia Sinica, Taiwan. She has published in the area of family sociology and youth studies. Dr Yi is an active member of the ISA and also serves in the editorial board for several sociological journals.

Wei-Chun Tsao is a research assistant at the Institute of Ethnology, Academia Sinica, Taiwan. She was editorial assistant of Formosa Journal of Mental Health (2008-2010). Her research interests are interdisciplinary with a focus on methodology of social sciences.

Po-San Wan is a research officer at the Hong Kong Institute of Asia-Pacific Studies, Chinese University of Hong Kong. She has published a number of books and has contributed to such jour- nals as International Political Science Review and Social Indicators Research.

Résumé

Cet article étudie quelles sont, à l’époque actuelle, les fonctions et les implications de la piété fil- iale dans trois sociétés chinoises, à savoir Taïwan, Hong-Kong et la Chine, à partir d’un vaste ensemble de données transnationales tirées du East Asian Social Survey de 2006. Malgré les valeurs culturelles confucéennes que partagent ces trois sociétés, elles se sont nettement différen- ciées tout au long du siècle dernier, tant au niveau des voies qu’elles ont empruntées vers la mod- ernisation que du développement de leurs structures sociopolitiques. Les auteurs pensent que les implications et les influences de la piété filiale ont tendance à présenter plus de similitudes à Taïwan et à Hong-Kong qu’en Chine, en raison de son système sociopolitique profondément dif- férent. En utilisant le modèle de piété filiale double comme référence pour des analyses compara- tives, les résultats montrent que la piété filiale double peut être observée dans les trois sociétés, quoique avec un certain nombre d’altérations componentielles en Chine. L’étude va également au-delà de la pratique courante qui consiste à limiter la piété filiale à l’attention portée aux parents plus âgés, en prenant en considération sa capacité à influer sur les aspects socio-psychologiques individuels. Les résultats de la régression tendent à confirmer l’importance de la piété filiale dou- ble et son étroite corrélation avec différents aspects de la vie quotidienne dans les sociétés chi- noises contemporaines.

Mots-clés: Culture chinoise, East Asian Social Survey, étude comparative, modèle de piété filiale double

Resumen

Este trabajo investiga las funciones e implicaciones de la devoción filial contemporánea en tres sociedades chinas, a saber, Taiwán, Hong Kong y China, usando bases de datos trasnacionales a gran escala de la East Asian Social Survey de 2006. A pesar de los valores culturales confucionis- tas compartidos por estas tres sociedades, las mismas se han diferenciado profundamente en su camino hacia la modernización y en el desarrollo de sus estructuras sociopolíticas a lo largo del siglo pasado. Los autores proponen que las implicaciones e influencias de la devoción filial tienden a ser más similares entre Taiwán y Hong Kong, al tiempo que son diferentes en China como con- secuencia de las profundas diferencias en su sistema sociopolítico. Usando el modelo dual de la devoción filial como modelo base para el análisis comparativo, los resultados muestran que la devoción filial dual puede ser encontrada en las tres sociedades, aunque hay algunas alteraciones

296 International Sociology 28(3)

en cuanto a sus componentes en China. El estudio también va más allá de la práctica habitual de tratar la devoción filial dentro de los confines del cuidado de los mayores de la familia al consid- erar su capacidad para influir en los aspectos socio-psicológicos individuales. Los resultados de la regresión muestran la significación de la devoción filial dual y su estrecha asociación con varios aspectos de la vida diaria en las sociedades chinas contemporáneas.

Palabras clave: Cultura china, East Asian Social Survey, estudio comparativo, modelo dual de la devoción filial