Attachment between infants and mothers in China

Abstract

The first distribution of Chinese infant–mother (n = 61) attachment classifications categorised by trained and reliability-tested coders is reported with statistical comparisons to US norms and previous Chinese distributions. Three-way distribution was 15% insecure-avoidant, 62% secure, 13% insecure-resistant, and 4-way distribution was 13% insecure-avoidant, 58% secure, 16% insecure-resistant,13% disorganised. These findings support the hypotheses that: (1) consistent with global norms the majority of infants will show secure attachments to mother; (2) insecure-resistant attachments will be greater than insecure-avoidant attachments in this interdependent cultural setting; and (3) disorganised attachments will be comparable to Western norms. Pooled samples from previously reported Chinese samples demonstrate deviations from US norms on all categories including relatively low proportions of avoidant and disorganised classifications, especially among samples from the South. Culture-specific childrearing practices and the role of training and reliability-testing for coders are suggested as possible contributors to these differences.

Keywords

Infant–mother attachment China Chinese children avoidant resistant secure disorganised

Introduction

The ‘gold standard’ of attachment research, the Strange Situation Procedure (SSP; Ainsworth, Blehar, Waters & Walls, 1978), has been implemented in dozens of different cultural settings almost always yielding similar distributions of infant–mother attachment across the four main patterns, i.e. secure, avoidant, resistant, and disorganised (see van IJzendoorn & Sagi-Schwartz, 2008 for a review). However, while China has a vast population and pronounced emphasis on family relationships, the few SSP studies conducted there have mostly been reported in the Chinese language. This paper reports on the first ever distribution of Chinese infant–mother attachment classifications derived from trained and reliability-tested coders, provides a summary of findings from previous Chinese studies, and presents findings of statistical tests for inter and intra-cultural differences.

Chinese cultural context and childrearing

Confucian-heritage cultures are traditionally composed of vertical collectives, wherein individual conduct is role-based, and relations to others accord with a prescribed hierarchy in which submission to authority is valued and expected (Chao, 1994; Ho, 1986; Kim, 2000; Ko, 2004). Family cohesion is a first point of reference in the conduct of all members – with filial piety an essential element – and thus, the pairing of partners, the timing of reproduction, and childrearing are fundamentally shared responsibilities with hierarchical guidance (Chen, Bond & Tang, 2007; Han, 2010; Ho, 1996; Liu, 2008).

This emphasis on interdependence has been observed and documented within the infant–mother dyad (Keller et al., 2009). For example, when compared to US mothers, not only did Chinese mothers of 3-month-olds endorse family allocentricism more strongly, making greater mention of body-contact behaviours (associated with relatedness promotion) and less mention of object-stimulation (associated with autonomy promotion), they also used less autonomy promoting verbal discourse during play (Keller, Abels, Borke, Lamm, Su & Wang, 2007). Among samples of 2-year-olds in Canada and China, Liu, Chen, Rubin, Zheng, Cui, Li, Chen, and Wang (2005) found that Chinese mothers were significantly more involved in mother–child interactions (encouraging both more connected and autonomous behaviours) during a 10-minute free play episode, and that Chinese infants displayed significantly more connectedness and significantly less autonomy than did the North American infants.

In addition to being short, the chords between infants and caregivers in contemporary China are also numerous and overlapping; childrearing almost invariably involves multiple-caregiver networks, often living in 3-generation households (Hu & Meng, 1996; Zhang, 2004). It is also common practice for mothers to return to work within 3 months of giving birth and for grandparents to assume primary caregiving responsibility. The current reproductive generation (including the parents in our sample) are for the most part products of the ‘One Child Policy’, and so are in the middle of what is known as the ‘4-2-1’ problem: one couple must care for 4 parents, and their own infant is the sole progeny of 6 adults (Hesketh, Lu, & Zing, 2005; Wang & Fong, 2009). Studies have found negative outcomes associated with both 3-generation household arrangements (reported in Miao & Wang (2003), citing Chen (1994, 2001)), and the ‘little emperor syndrome’ (e.g. Wu, 1996). While the present study does not include direct measures of indulgence, it does consider household structure and density of caregiver network variables in relation to attachment classification.

The Strange Situation Procedure and cultural context

Applying the SSP in different cultural settings may yield different distributions of attachment classification, possibly linked to differences in prevailing caregiving attitudes and behaviours. It has been argued, for example, that the high proportions of resistant classifications found in Japan and Korea are the result of either/both too great a distress in SSP separations (for a culture in which infants are rarely left alone by mother) and/or are manifestations of culturally-valued traits of dependency and indulgence within the dyad (e.g. Jin, Jacobvitz, Hazen & Jung, 2012; Rothbaum, Kakinuma, Nagaoka & Azuma, 2007;Takahashi, 1986). High proportions of insecure-avoidant classifications found in an early German sample were, conversely, explained in terms of a cultural context which highly values individualism and independence (Grossmann, Grossmann, Spangler, Suess & Unzner, 1985). However, there remains a conspicuous gap between the posited culture-specific theoretical frameworks and the empirical evidence, particularly with other researchers presenting contrary results such as ‘normative’ proportions of insecure sub-classifications in Japan (Behrens & Kondo-Ikemura, 2011; Durett, Otaki & Richards, 1984). More data from distinctly independent and interdependent cultures are needed, and the present study contributes to this endeavour.

Previous SSP-based research in China and the present study

Access to and engagement with international researchers and trainers has been a notable challenge in extending attachment research to China. Personal communications with Chinese developmental psychologists including Hu Ping and Ding Yan Hua, researchers on, respectively, the earliest (1996) and most recent (2012) SSP studies outlined below, have confirmed that there are language and geographical barriers. As far as we are aware only two SSP-based studies with Chinese dyads have been published in English (Ding, Xu, Wang, Li & Wang, 2012; Trnavsky, 1998), neither of which included coding by trained and reliability-tested judges. Seven further SSP-based studies conducted in China have been identified, six of which are included in the analyses with our new sample (Ding, Wang, Li, Chi & Xu, 2008; Gu, Cen, Li, Gao, Li & Chen, 1997; Li, Jing, Yang, Cai, Chen & Su, 2004; Liang, Chen & Chen, 2000; Hu & Meng, 2003; Yue, Zhang, Chen, Liang & Zhang, 2010). Sample sizes and distribution frequencies are provided in Table 1.

Table 1.

Comparison of attachment classification distributions between US norms and new Chinese sample, previous Chinese samples, and Northern/Southern pooled samples (3-way, 4-way, with percentages and adjusted standardised residuals).

		Distributions					Adjusted Standardized Residuals
	n	A	B	C	D	χ²	A	B	C	D
3-way
US norm¹	1584	21%	67%	12%	–
New Chinese sample	61	14.8%	62.3%	23%	–	7.43*	−3.8*	−2.9	6.7*	–
Hu & Meng, 1996	31	16.1%	67.7%	16.1%	–	0.79	−1.5	0.2	1.3	–
4-way
US norm²	2104	14.8%	61.7%	8.7%	14.8%
New Chinese Sample	61	13.1%	57.4%	16.4%	13.1%	4.57	−1.0	−2.6	4.7	−1.0
Trnavsky, 1998	29	24.1%	65.5%	10.3%	–	4.21	2.6	0.5	0.4	−3.4
Liang et al., 2000	122	10.7%	73%	7.4%	9%	6.92	−5.1	13.7	−1.6	−7.1
Hu & Meng, 2003 (A)	30	13.3%	53.3%	26.7%	6.7%	12.86**	−0.4	−2.5	5.4*	−2.4
Hu & Meng, 2003 (B)	34	14.7%	52.9%	20.6%	11.8%	6.16	0.0	−3.0	4.0	−1.0
Yue et al., 2010	178	10.1%	71.9%	9.6%	8.4%	10.68**	−8.3*	18.2*	1.5	−11.3*
Gu et al., 1997	52	5.8%	67.3%	23.1%	3.8%	19.7***	−4.7*	2.9	7.5*	−5.7*
Li et al., 2004	75	17.3%	65.3%	13.3%	4%	8.25*	1.9	2.7	3.5*	−8.1*
Ding et al., 2008	62	6.5%	67.7%	25.8%	–	30.91***	−5.3*	3.1	10.5*	−8.3*
Ding et al., 2012	160	7.5%	68.1%	21.9%	2.5%	55.11***	−11.7*	10.3*	21.1*	−19.7*
Northern total:	454	12.1%	67.2%	11.9%	8.8%	19.8***	−11.2*	27.5*	9.8*	−26.2*
Southern total:	349	9.2%	67.3%	20.9%	2.6%	104.37***	−19.7*	19.7*	42.6*	−42.7*

Note. Attachment Classifications: A = Insecure-Avoidant, B = Secure, C = Insecure-Resistant, D = Disorganised. Source of Norms: ¹ van IJzendoorn et al., 1992, ² van IJzendoorn et al., 1999. *p < .05. **p < .01. ***p < .001.

While it is the largest known Chinese SSP-based study – including 187 14-30 months olds – Deng, Zou, Jin, Tang, Li, Cen and Zou’s (2007) sample is excluded from analyses as it assessed autistic infants, finding only 8% to have a secure attachment, 67% insecure-avoidant and 23% disorganised. It is of particular interest that among the 9 included samples disorganised classifications are notably rare, with an average prevalence of just 5.5%.

China comprises an immense population spread over a vast area, and so we might expect internal – or intra-cultural – variations. Such has been the case within much smaller countries (for example, fewer avoidant classifications in Southern Germany, Escher-Graeub & Grossmann, 1983, cited in Grossman et al., 1985)). Recent research supports a North–South cultural divide in China, with those raised in the North demonstrating a significantly greater tendency toward an independent-type understanding of relationships (Talhelm et al., 2014). These authors suggest that this pattern, which they have labelled the ‘Rice Theory’, stems from the marked difference in cooperation and interdependence within the extended social group required to cultivate rice in the South, compared to wheat in the North. In order to detect any effect that this difference might have on childrearing and resulting infant–caregiver attachment patterns, the current article includes comparative analyses of samples from the North and South.

As of 2010 official figures showed that more than half of China’s 1.3 billion citizens were normally situated in urban areas (approximately 47% urban residents, plus 16% ‘floating’ workers who migrate to more affluent areas for work but are not granted access to urban resources (National Population and Family Planning Commission of P.R.C., 2010)). As with the previous Chinese studies cited here, the present study focuses on this majority and rapidly expanding urban population, the household characteristics of which are becoming increasingly representative of the population at large.

This article hypothesised that among its new Chinese sample: (1) consistent with the overwhelming trend among findings in diverse cultural settings the majority of infants will be found to be securely attached; and (2) consistent with an established trend among interdependent cultures, insecure-resistant classifications will be more prevalent than insecure-avoidant classifications. Next, in light of the lack of training among researchers on previous studies, and the acknowledged difficulty of coding for its indicators (van IJzendoorn, & Bakermans-Kranenburg (1999)), we expect that (3) the prevalence of disorganised attachment classifications will not differ significantly from Western norms. Finally, given the long recognised and recently evidenced North–South cultural division in this immense country, we predict that (4) the proportion of insecure-resistant classifications will be significantly greater among samples from the South of China where interdependence is more pronounced.

Method

Participants

One of the primary objectives of this study was to provide a comparison sample for an attachment study of children raised in Chinese orphanages (Steele et al., 2009; publications to follow). Drawn from a pre-K group at a kindergarten affiliated with North-West China’s Shaanxi Normal University in the provincial capital Xi’an, the sample consists of 61 children (57% female) aged between 12 and 38 months (average 20.8 months) and their mothers (age range 24–38 years, average 30.7 years).

Measures and procedure

The widely-used and reported-upon SPP was used to evoke infant–caregiver attachment behaviours. Participants were assigned a 1-hour slot in advance of the test day, at which time a Chinese research assistant led them from the first-floor pre-K activity room to the laboratory set-up on the second floor. Care was taken to ensure infants were reasonably calm and comfortable in an office adjacent to the testing room, while the mother was instructed on the SSP and completed informed consent forms. The research assistant then introduced the dyad to the unfamiliar room and served as timekeeper, with a second female Chinese research assistant acting as ‘stranger’.

Household Structure, Childcare Practices, and other demographic details were recorded immediately following the SSP, through a brief structured interview, conducted by a Chinese research assistant.

Coding and analysis of data

SSP coding and categorisation into 3-way and 4-way distributions was performed according to detailed criteria by reliability-tested coders (Ainsworth et al., 1978; Main & Solomon, 1986, 1990). All recorded SSPs were rated by FH, trained by Alan Sroufe and Elizabeth Carlson. One-quarter (16/61) of the procedures were also coded by a second reliable and highly experienced coder HS (initially trained in 1988 by Mary J. Ward), with 88% agreement (Kappa = .80). Responses to brief structured interview items were translated into English by the Chinese research assistant and the first author, immediately after procedures were completed.

Utilising an approach well established in SSP meta-analyses (e.g. van IJzendoorn & Kroonenberg, 1988), we employed multinomial tests to compare our sample distribution, and those for all known previous SSP studies, to US norms (van IJzendoorn, Goldberg, Kroonenberg & Frenkel (1992) for the 3-way norm; van IJzendoorn et al. (1999) for the 4-way norm; see Table 1). Due to the large number of tests included, and following a recent and comparable examination of Italian SSP sample distributions (Cassibba, Sette, Bakermans-Kranenburg & van IJzendoorn, 2013), standardised residuals exceeding 3.29 (or -3.29, corresponding to a p-value of .001) are deemed indicative of a significant deviation from the norm.

Results

Given the wide age-range of the children in the sample, compared to that for which the SSP is best-validated (and the established norms against which this sample will be compared), the distribution among those in the 12–18 month age-range was compared to those in the 19–38 month age-range. No statistically significant difference was found (3-way, χ² (2, n = 61) = 2.06, p = .36; 4-way, χ² (3, n = 61) = 1.97, p = .58), and so the whole sample was used in the main analyses.

The 3-way distribution for our new sample is: A = 14.8%, B = 62.2%, C = 23%, which differs significantly from the US norm [χ² (2, n = 1,645) = 7.43, p = 0.02]. Examination of the adjusted residuals shows that our Chinese sample contains significantly more insecure-resistant classifications (23% vs 12%, z = 6.7) and fewer insecure-avoidant classifications (14.8% vs 21%, z = −3.8). By contrast, the single previously reported 3-way distribution with a Chinese sample (Hu & Meng, 1996) did not differ significantly to the US norm [χ² (2, n = 1,615) = 0.79, p = .7]. However, there was no significant difference between our sample and this previously reported sample [χ² (2, n = 92) = 0.58, p = .75].

The 4-way distribution for our new Chinese sample is: A = 13.1%, B = 57.4%, C = 16.4%, D = 13.1%, which does not differ significantly from the (van IJzendoorn et al., 1999) US norm [χ² (3, n = 2165) = 4.57, p = .21]. By contrast, 6 of the previous 9 Chinese sample distributions differ significantly from the US norm [Hu & Meng, 2003 (A), χ² (3, n = 2,135) = 12.86, p = .007; Yue et al., 2010, χ² (3, n = 2, 282) = 10.68, p = .02; Gu et al., 1997, χ² (3, n = 2, 156) = 19.7, p < .001; Li et al., 2004, χ² (3, n = 2,179) = 8.25, p = .04; Ding et al., 2008, χ² (3, n = 2, 166) = 30.91, p < .001; Ding et al., 2012, χ² (3, n = 2, 264) = 55.11, p < .001]. Notably, all four samples from the South of China and just two of the five samples from the North of China differed to the US norm. Inspection of adjusted residuals showed that for the Southern samples, 12 of the total 16 classification cell values (4 categories ×4 samples) deviated significantly, compared to just 4 of 20 classifications among the Northern samples (full data provided in Table 1).

Pooling samples by region (with our new sample included in the Northern group) yielded the following distributions: Northern (6 samples, n = 454): A = 12.1%, B = 67.2%, C = 11.9%, D = 8.8%; Southern (4 samples, n = 349): A = 9.2%, B = 67.3%, C = 20.9%, D = 2.6%. Both Northern and Southern distributions differed significantly from the US norm [Northern χ² (3, n = 2,558) = 19.8, p < .001; Southern: χ² (3, n = 2,453) = 104.37, p < .001], with significant deviations in every category (see Table 1). In both cases there were significantly more secure and insecure-resistant classifications, and fewer insecure-avoidant and disorganised classifications, compared to US norms.

These Northern and Southern total distributions also differ significantly from one another [χ² (3, n = 803) = 24.30, p < .001]. Inspection of the adjusted residuals confirmed that insecure-resistant classifications are proportionately higher among the Southern group (z = 3.6), while disorganised classifications are proportionately higher among the Northern group (z = 3.7; see Table 2). This regional difference was maintained when our new Chinese sample was excluded from the Northern group [χ² (3, n = 2,497) = 22.84, p < .001].

Table 2.

Comparison of attachment classification distributions between Northern (including our New Chinese sample) and Southern samples pooled into groups (3-way, 4way, with percentages and adjusted standardised residuals).

		Distributions					Adjusted Standardized Residuals
	n	A	B	C	D	χ²	A	B	C	D
Northern total:	454	12.1%	67.2%	11.9%	8.8%					−
Southern total:	349	9.2%	67.3%	20.9%	2.6%	24.3***	−1.3	.0	3.5*	−3.7*

Note. Attachment Classifications: A = Insecure-Avoidant, B = Secure, C = Insecure-Resistant, D = Disorganised. *p < .05. **p < .01. ***p < .001.

Finally, comparing our sample to all previous Chinese samples pooled as a single group (n = 742, distribution: A = 10.6%, B = 68.1%, C = 15.8%, D = 5.5%) found no statistically significant difference [χ² (3, n = 803) = 6.61, p < .09].

Household structure and childcare practices

Overall, 64% (n = 38) of infants were reported to spend most time with someone other than mother, and 29% (n = 17) of infants were reported to be closer to someone other than mother (usually grandmother in both cases). Half of all infants (49.2%, n = 29) lived with parents in 2-generation households, with the rest living in 3-generation households. 32% of infants (n = 19) lived with four adults, 17% (n = 10) lived with 3 adults, and 1 infant lived with 6 adults. Number of adults in household is not a definite indicator of the caregiver-to-infant ratio, as typically experienced by the child, because in 43% (n = 13) of 2-generation households a grandparent was identified as spending most time with infant (compared to 86%, n = 25, in 3-generation households). Only 14% (n = 8) of infants had experienced time in day-care. Mothers in this middle class urban sample were all married, and most (85%) were university graduates. 95% of infants (n = 56) were only children, and 13.5% (n = 8) had neither cousins nor siblings. Chi square tests found that none of these was significantly associated with attachment classification.

Discussion

As expected the majority of our sample was found to be securely attached consistent with global norms. Moreover, this was the case for all previously reported Chinese samples, of which only three included prevalence of secure classifications that differed significantly from US norm in 4-way distributions (adjusted residual z-score above 3.29), and in all cases secure attachments were more abundant.

Consistent with our second prediction, we found relatively more insecure-resistant than insecure-avoidant attachments. While the overall distribution in our sample did not differ significantly from the US norm, the z-score for proportion of resistant classification indicated a significant deviation. Given the age range, reports of most children spending time with another caregiver, and experience of the pre-K program, it would seem improbable that –as has been suggested for Japanese infants – insecure-resistant behaviours were evoked as a result of unfamiliar distress in the SSP.

In support of our third hypothesis, the prevalence of disorganisation among our sample was extremely close to the US norm; while the 4-way distribution did not differ significantly from the pooled group of previous Chinese samples, it approached significance, and prevalence of disorganised classifications was markedly higher. Five of the nine previous Chinese studies included significantly lower percentages of disorganised classifications compared to the US norm. We take this as tentative support for our supposition that disorganised behaviours and attachment patterns have previously been under-detected due to lack of coder training in China. This finding requires replication but suggests that disorganisation is no less ‘culturally’ common, and provides a reference for future research with higher-risk populations, including our investigation of infant–caregiver attachment in Chinese institutions.

In support of our last hypothesis, the Southern group differed significantly to the Northern group with a higher proportion of insecure-resistant classifications, and individual Southern samples showed similar and more pronounced deviations from the US norm. This overall pattern is consistent with the suggestion that, not only are infant–mother attachment relationships within Chinese families characterized by a greater tendency toward insecure-resistant type interactions, but this tendency is more pronounced in the South. Extending the postulated association between insecure-resistant infant–caregiver attachments and a cultural background that emphasises interdependence, this finding is also consistent with evidence that individuals who are raised in the South of China demonstrate a significantly greater tendency toward interdependent-type understandings of relationships (Talhelm et al., 2014).

That the new sample presented here, which is drawn from Northern China, included a relatively large proportion of insecure-resistant classifications is a notable anomaly. One possible explanation for this could be that this is the only sample from China’s Western interior, which might present a further line of cultural distinction. Alternatively, it might be the case that the higher frequency of insecure-resistant-like behaviours is associated with the material and caregiver provision experienced by infants in this upper middle-class and highly educated sample. However, while the percentage of mothers who had graduated from university was very high – 85% compared to an average of around 30% among first-tier city residents (Burkholder & Lyons, 2005), which is likely to be closer to the figure among samples from previous studies – this variable was not found to be associated with infant-caregiver attachment classification.

Clearly though, the study sample and all of those included in the reports, are drawn from an urban middle class which may be more comparable to those in similar groups of other nations than their rural counterparts. While the urban population is rapidly expanding, and urban residency (not including migrant workers temporarily located in cities) accounts for almost half of the total, given the immensity of the population and the likely shifts in caregiving practices with modernisation, it is vital that research be extended to cover diverse and minority groups. Considering recent findings within Italian populations changes over time as well as space might also be predicted, with more insecure-avoidant behavioural patterns emerging in tandem with a shift toward greater resource security and independence of family members. Indeed, a ‘Value of Children’ study across several countries found that while rural Chinese mothers, and those who migrated temporarily from rural to urban areas for work, endorsed family models of interdependence those from urban settings tended toward family models of independence similar to mothers from Western-urban samples (Zheng & Shi, 2004). SSP studies within rural, migrant, and dispersed families are therefore called for. Besides deepening our understanding of anticipated cultural differences, such research might identify risks associated with children who are displaced or ‘left-behind’ when their parents migrate for work.

Because this study sample included 95% only-children, comparisons to those with siblings were not possible. However, the broadly ‘normal’ distribution suggests no detrimental impact on attachment; moreover, the absence of gender differences may support suggestions that traditional preferential treatment of boys has waned as a result of the one-child policy (Hesketh, Lu & Xing, 2005; van IJzendoorn & Sagi-Schwartz, 2008; Qu & Hesketh, 2006).

That almost two-thirds of our sample spent most time with someone other than their mother is consistent with survey reports in other major cities, which found up to 89% of children were cared for primarily by grandparents (Honggen (2007), cited in Settles, Sheng, Zang & Zao (2013)). That neither this experience, nor being raised in a 3-generation household, was associated with infant–mother attachment is important given concerns that such arrangements might be detrimental to infant development (e.g. Wu, 1996). Future research should address diverse interactions within the social caregiver network, attending particularly to the role of grandmothers and the balance of caregiving within the mother–grandmother relationship (van IJzendoorn, 2005).

We applaud the pioneering contributions of previous Chinese researchers and, to enhance comparability and innovation, support ongoing work with an increased exchange of ideas, methods, and findings. Re-examination of existing tapes from previous studies by reliability-tested coders, and further collaboration between local and Western researchers on theory and methodology, are also suggested as avenues for future investigation.

The findings of this report have increased accessibility to little-known previous findings, and provided the first distribution coded by reliability-tested researchers. On balance, the findings presented add support to the widely demonstrated statistical ‘normativity’ of secure attachments, suggest the presence of both inter- and intra- cultural tendencies toward insecure-resistant attachments, and identify typical levels of disorganisation within a Chinese sample.

Footnotes

Acknowledgments

We thank the reviewers of an earlier draft of this paper, and Marinus van IJzendoorn, for guidance on treatment of data and suggestions for improving the text.

Funding

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

References

Ainsworth

M. D. S.

Blehar

M. C.

Waters

Walls

(1978). Patterns of attachment: a psychological study of the strange situation. Hillsdale, NJ: Lawrence Erlbaum.

Behrens

K. Y.

Kondo-Ikemura

(2011). Japanese children’s amae and mothers’ attachment status as assessed by the Adult Attachment Interview. International Journal of Psychology, 46(5), 368–376. doi: 10.1080/00207594.2011.561854.

Burkholder

Lyons

(2005). Education in China: Availability, Duration, and Quality Improving. Summary of Gallup Survey Report. Available at: http://www.gallup.com/poll/14932/education-china-availability-duration-quality-improving.aspx (accessed 8 September 2014).

Cassibba

Sette

Bakermans-Kranenburg

M. J.

van IJzendoorn

M. H.

(2013). Attachment the Italian way: in search of specific patterns of infant and adult attachments in Italian typical and atypical samples. European Psychologist, 18(1), 47–58.

Chao

R. K.

(1994). Beyond parental control and authoritarian parenting style: understanding Chinese parenting through the cultural notion of training. Child Development, 65(4), 1111–1119.

Chen

H. C.

(1994). The characteristics of influencing factors and measurement of child social development. Psychological Development and Education, 4, 1–16.

Chen

S. X.

Bond

M. H.

Tang

(2007). Decomposing filial piety into filial attitudes and filial enactments. Asian Journal of Social Psychology, 10, 213–223.

Deng

Zou

X. B.

Jin

Tang

Cen

Zou

(2007). Chinese Journal of Clinical Psychology, 15(5), 480–482.

Ding

Y. H.

Wang

Z. Y.

H. R.

Chi

Z. Y.

(2008). Study of 62 infants’ attachment in Shanghai and their relationship with infants’ temperament and cognitive development. Chinese Child Healthcare Journal, 16(2), 163–165.

10.

Ding

Wang

Z.Y

Wang

(2012). Study of mother–infant attachment patterns and influence factors in Shanghai. Early Human Development, 88, 295–300.

11.

Durett

M. E.

Otaki

Richards

(1984). Attachment and the mother’s perception of support from the father. International Journal of Behavioural Development, 7, 167–176.

12.

Grossmann

K. E.

Spangler

Suess

Unzner

(1985). Maternal sensitivity and newborns’ orientation responses as related to quality of attachment in Northern Germany. Monographs of the Society for Research in Child Development. Growing Points of Attachment Theory and Research, 50(1/2), 233–256.

13.

Cen

Gao

Chen

(1997). Two-year-old children’s social behaviour development and related family factors. Psychological Science, 20, 519–524.

14.

Han

(2010). Trends in educational assortative marriage in China from 1970 to 2000. Demographic Research, 22, 733–770.

15.

Hesketh

Xing

Z. W.

(2005). The effects of China’s One-Child Family Policy after 25 years. New England Journal of Medicine, 353, 1171–1176.

16.

D. Y. F.

(1986). Chinese patterns of socialisation: A critical review. In Bond

M. H.

(Ed.), The Psychology of the Chinese People (pp. 1–37). Hong Kong: Oxford University Press.

17.

D. Y. F.

(1996). Filial piety and its psychological consequences. In Bond

M. H.

(Ed.), The Handbook of Chinese Psychology (pp. 155–165). Hong Kong: Oxford University Press.

18.

Meng

(1996). An examination of infant–mother attachment in China. Poster presented at the meeting of the International Society of the Study of Behavioral Development, Quebec City, Quebec, Canada.

19.

Meng

(2003). Research on discrimination of mother–infant attachment type. Acta Psychologica Sinica, 35, 201–208.

20.

Jin

M. K.

Jacobvitz

Hazen

Jung

S. H.

(2012). Maternal sensitivity and infant attachment security in Korea: cross-cultural validation of the Strange Situation. Attachment & Human Development, 14(1), 33–62.

21.

Keller

Abels

Borke

Lamm

W. L. Y.

Wang

(2007). Socialization environments of Chinese and Euro-American middle-class babies: Parenting behaviors, verbal discourses and ethnotheories. International Journal of Behavioral Development, 31(3), 210–217.

22.

Keller

Borke

Staufenbiel

Yovsi

R. D.

Abels

Papaligoura

… Su

(2009). Distal and proximal parenting as alternative parenting strategies during infants’ early months of life: a cross-cultural study. International Journal of Behavioral Development, 33, 412–420.

23.

Kim

(2000). Indigenous, cultural, and cross-cultural psychology: A theoretical, conceptual, and epistemological analysis. Asian Journal of Social Psychology, 3, 265–287.

24.

N. H.

(2004). Familism in Confucianism. Paper presented at: The International Conference of Women’s global connection, San Antonio, Texas, USA.

25.

Jing

Yang

Cai

Chen

(2004). Characters of 75 infants’ attachment towards their mothers. Chinese Mental Health Journal, 18(5), 291–293.

26.

Liang

Chen

(2000). Pattern of toddler–mother attachment. Psychological Science, 23(3), 324–328.

27.

Liu

(2008). Negotiating the Filial Self: Young-Adult Only-Children and Intergenerational Relationship in China. YOUNG: Nordic Journal of Youth Studies, 16(4), 409–430.

28.

Liu

Chen

Rubin

K. H.

Zheng

Cui

Chen

Wang

(2005). Autonomy- vs. connectedness-oriented parenting behaviors in Chinese and Canadian mothers. International Journal of Behavioral Development, 29, 489–495.

29.

Main

Solomon

(1986). Discovery of an insecure-disorganized/disoriented attachment pattern: Procedures, findings, and implications for the classification of behavior. In Brazelton

T. B.

Yogman

(Eds.), Affective development in infancy (pp. 95–124). Norwood, NJ: Ablex.

30.

Main

Solomon

(1990). Procedures for identifying disorganized/disoriented infants in the Ainsworth Strange Situation. In Greenberg

M. T.

Cicchetti

Cummings

E. M.

(Eds.), Attachment in the preschool years: Theory, research, and intervention (pp. 121–160). Chicago: University of Chicago Press.

31.

National Population and Family Planning Commission of P.R.C. (2010). Total Population by Urban and Rural Residence and Birth Rate, Death Rate, Natural Growth Rate by Region, 2009. Available online at: http://www.npfpc.gov.cn/ (accessed 21 December 2010).

32.

J. D.

Hesketh

(2006). Family size, fertility preferences, and sex ratio in China in the era of the one child family policy: results from a national family planning and reproductive health survey. British Medical Journal, 333, 371–373.

33.

Rothbaum

Kakinuma

Nagaoka

Azuma

(2007). Attachment and Amae: parent–child closeness in the United States and Japan. Journal of Cross-Cultural Psychology, 38(4), 465–486.

34.

Settles

B. H.

Sheng

X. Zang Y.

Zhao

(2013) The one child policy and its impact on Chinese families. In Kwok-bun

Chan

(Ed.), International handbook of Chinese families (pp. 627–646). New York: Springer.

35.

Steele

Jin

Archer

Herreros

(2009). Effects of lessening the level of deprivation in Chinese orphanage settings: Decreasing disorganization and increasing security. Paper presented at: The Biennial Meeting of the Society for Research in Child Development. April 2-4, 2009; Denver, CO.

36.

Takahashi

(1986). Examining the strange-situation procedure with Japanese mothers and 12-month-old infants. Developmental Psychology, 22, 265–270.

37.

Talhelm

Zhang

Oishi

Shimin

Duan

Lan

Kitayama

(2014). Large-scale psychological differences within China explained by rice versus wheat agriculture. Science, 344, 603–608.

38.

Trnavsky

(1998). Strange situation behaviors in Chinese infants. Child Study Journal, 28, 1–13.

39.

van IJzendoorn

M. H.

(2005). Attachment in social networks: toward an evolutionary social network model. Human Development, 48, 85–88.

40.

van IJzendoorn

M. H.

Kroonenberg

P. M.

(1988). Cross-cultural patterns of attachment: a meta-analysis of the strange situation. Child Development, 59(1), 147–156.

41.

van IJzendoorn

M. H.

Goldberg

Kroonenberg

P. M.

Frenkel

O. J.

(1992). The relative effects of maternal and child problems on the quality of attachment: a meta-analysis of attachment in clinical samples. Child Development, 63, 840–858.

42.

van IJzendoorn

M. H.

Schuengel

Bakermans-Kranenburg

M. J.

(1999). Disorganized attachment in early childhood: meta-analysis of precursors, concomitants, and sequelae. Development and Psychopathology, 11, 225–249.

43.

van IJzendoorn

M. H.

Sagi-Schwartz

(2008). Cross-cultural patterns of attachment: universal and contextual dimensions. In Cassidy

Shaver

P. R.

(Eds.), Handbook of attachment: theory, research, and clinical applications (pp. 713–734). Second edition. New York: Guilford Press.

44.

Wang

Fong

V. L.

(2009). Little Emperors and the 4:2:1 Generation: China’s Singletons. Journal of the American Academy of Child and Adolescent Psychiatry, 48(12), 1137–1139.

45.

D. Y. H.

(1996). Chinese Childhood Socialization. In Bond

M. H.

(Ed.), The handbook of Chinese psychology (pp. 141–154). Hong Kong: Oxford University Press.

46.

Yue

Zhang

Chen

Liang

Zhang

(2010). The relations between children’s attachment and problem behaviour. Psychological Science, 33(2), 318–320.

47.

Zhang

Q. F.

(2004). Economic transition and new patterns of parent–adult child coresidence in urban China. Journal of Marriage and Family, 66(5), 1231–1245.

48.

Zheng

Shi

(2004). Intercultural and Intracultural Differences in the Value of Children: Comparisons between Four Countries and the Urban, Rural, and Floating Populations in China. In Zheng

Leung

Adair

(Eds.), Perspectives and Progress in Contemporary Cross-Cultural Psychology: Selected Papers from the Seventeenth International Congress of the International Association for Cross-Cultural Psychology (Online Edition). Retrieved from http://iaccp.org/ebook/xian/chapters/zheng.php