ArticleInfant–care provider attachments in contrasting child care settings I: Group-oriented care before German reunification
Introduction
Little is known about the factors affecting the formation of attachments between infants and professional child care providers. Attachment relationships are believed to be grounded in the interactional histories of infants, but it remain unclear how dyadic- and group-level interactions in group settings affect the development of trusting relationships with nonfamilial providers. When infants are observed in laboratory situations with their mothers, care providers, and/or strangers, they typically prefer their mothers, but also direct more attachment behaviors (proximity seeking, contact seeking, calling, and following) to care providers than to strangers Barnas and Cummings 1994, Cummings 1980, Fox 1977. In child care settings, infants differentiate between stable and nonstable care providers, greeting stable care providers more warmly when they arrive at the center and turning to them more often during the day (Barnas & Cummings, 1994). In the presence of stable rather than nonstable care providers, infants cry less, are soothed more easily, and are more likely to turn to stable care providers when distressed. Such findings raise questions about the formation of infant–care provider attachments when, as in the former East Germany, care providers provided group- rather than infant-oriented care and emphasized peer rather than care provider–infant interactions. These issues were explored in the present study.
Before reunification in 1990, societal rather than individual responsibility for the care of children was emphasized in socialist East Germany (see, for discussion, Ahnert and Lamb 2001, Lamb and Sternberg 1992). Exclusive child care within the family was even denounced as a petit-bourgeois aspiration likely to have adverse effects on child development and on the establishment of social relationships (Schmidt-Kolmer & Schmidt, 1962). As in the other East block countries, child care authorities emphasized collectivism (Krupskaia, 1972), with children’s behaviors deemed appropriate only when they furthered group aims and goals. East German child care centers were thus expected to introduce group-oriented life experiences by fostering peer interactions rather than individual child–care provider relationships. The official curriculum instructed care providers to direct and control the infants’ behavior, even though warm relationships with the children were emphasized. As part of a larger educational system that included kindergartens, schools, and universities, children in child care were organized into age-homogenous groups. Well-organized educational instruction was thus provided to East German children from the very beginning of their lives, with central authorities licensing, administering, and establishing curricula for centers nationwide. Reflecting extreme beliefs about childhood plasticity, care providers were taught about group-oriented caretaking procedures, as well as how to play with and supervise individual infants. Conceptions of caretaking are deeply rooted in German culture, however, and reflect historical processes that began long before the 40 years of Communist rule. As a result, East German educational doctrines are unlikely to have reshaped the basic features of German childrearing in so brief a period of time (Ahnert et al., 1994).
In the present study, we sought to determine whether and how much the patterns of care observed before German reunification were group-oriented as opposed to individual-oriented, whether this affected the numbers of secure and insecure infant–care provider attachments relative to the numbers of secure and insecure infant-mother attachments, and whether both group and dyadic-level measures of care provider behavior predicted variations in the security of infant–care provider attachment.
An increasing number of researchers Ainslie 1990, Anderson et al 1981, Goossens and van IJzendoorn 1990, Sagi et al 1985, Sagi et al 1995 have used Ainsworth and Wittig’s (1969) Strange Situation to assess infant–care provider attachments, even though the measure was developed to explore variations in the patterning of attachment behaviors directed to those primarily responsible for infant care (Ainsworth et al., 1978). According to ethological attachment theorists, contact seeking in emotionally stressful situations is the hallmark of secure attachment Ainsworth et al 1978, Bowlby 1969, Bretherton 1991, Grossmann and Grossmann 1990, Lamb et al 1985, Sroufe and Waters 1977, and by this definition, it should be possible to identify secure attachments even to child care providers. By contrast, avoidance of those care providers in the Strange Situation is harder to interpret unless one can distinguish between avoidance and disinterest. Van IJzendoorn, Sagi, and Lambermon (1992) have thus proposed using indices of predictive or external validity (such as the quality of later peer relationships or care provider sensitivity) and estimations of the distribution across attachment categories when interpreting infant behavior toward child care providers in the Strange Situation. However, the distribution across attachment categories varies both intra- and interculturally (van IJzendoorn & Kroonenberg, 1988), indices of later peer relationships are seldom obtained, and associations between the sensitivity of child care providers and attachment security have been little studied. The operational implications of van IJzendoorn et al.’s (1992) suggestions are thus unclear.
Avoidance in the service of attachment may involve proximity to attachment figures despite the refusal to interact or seek contact immediately after reunion (Main, 1981). In the course of the Strange Situation, for example, avoidant infants (those in the A1 subgroup) might maintain proximity and self-regulate physiologically by socially referencing their attachment figures (Spangler & Grossmann, 1993). In addition, when the Strange Situation procedure is employed with preschoolers, who have more extensive repertoires of attachment behavior, social referencing may be one index of avoidant attachment Crittenden 1990, Crittenden 1992. Anticipating the possibility that we would observe many avoidant attachments because high rates of insecure-avoidant attachments have been found in other German samples (e.g., Gloger-Tippelt and Huerkamp 1998, Suess et al 1992, Grossmann et al 1985, Rottmann and Ziegenhain 1988), we thus used social referencing to help determine whether infants were avoidantly attached or not attached to their care providers Bradshaw et al 1987, Bretherton 1992, Camras and Sachs 1991, Klinnert et al 1986, Walden 1991. Specifically, we observed infants twice in the Strange Situation, with care providers playing the role of stranger in the first Strange Situation (before child care entry) and playing the role of mother in the second Strange Situation, five months later. We then compared the levels of social referencing, labeling infants “nonattached” to their care providers when they did not reference the providers more in the second Strange Situation than in the first (see Method section).
In the present study, we observed infants’ entry into child care and explored the period right after separation from the parents when a sensitive substitute was most likely to be sought. We assumed that the care providers’ attentiveness to individual children might be less pronounced and that less emotional support might be provided to individuals because the East German curriculum was group-oriented. Both factors would have made it more difficult for infants to form secure attachments to their care providers.
Section snippets
Infants
Forty healthy infants were observed prior to German reunification in 1987–1989. They were between 11 and 13 months old (M = 12.2 months, SD = 0.8) when they were enrolled in child care, and had been born at full-term. Their Bayley (1993) Mental Development Index (MDI) scores averaged 101.2 (SD = 17.2). Up to the time of enrollment in child care, all infants had been reared exclusively in East Berlin families, with their mothers as primary caretakers. A comparison with the sociological
Controlling for care provider behaviors
Because the 40 infants were cared for by ten care providers, we had to determine whether scores for the four infants observed with the same care provider could be treated as independent. We computed intraclass correlations (Winer, 1971) among scores for four dyads (the four children per care provider) on each of the six care provider behaviors observed 1 to 2 days, 2 to 4 weeks, and 3 to 4 months after the infants’ entry into child care. These estimates of the variance within the dyads over
Discussion
The research reported here began as a study of infant-mother and infant–care provider attachments in East Germany before reunification. At that time, the authorities disapproved of attachment theory, which they feared might undermine popular acceptance of nonmaternal care, even though paid maternity leaves of at least one year ensured that most infants only entered full-time child care in large age-homogeneous groups at around 12 months of age. In such circumstances, care providers were often
Acknowledgements
This research was financed by Ministry of Health in East Germany during the data collection, and by the Intramural Research Program at the National Institute of Child Health and Human Development in Washington/Bethesda, thereafter. We are grateful to the research group “Early Socialization” at Institute of Mental Health in Childhood and Adolescence in East Berlin, especially to B. Lillie and R. Fliedner, for data collection. We thank Karin and Klaus Grossmann (University of Regensburg) for
References (56)
Caregivers in daycare centersdoes training matter?
Journal of Applied Developmental Psychology
(1989)- et al.
Caregiver stability and toddlers’ attachment-related behavior towards caregivers in day care
Infant Behavior and Development
(1994) - et al.
Attachment, temperament, and social referencinginterrelationships among three domains of infant affective behavior
Infant Behavior and Development
(1987) - et al.
Social referencing and caretaker expressive behavior in a daycare setting
Infant Behavior and Development
(1991) - Ahnert, L., & Lamb, M. E. (2000). Infant–care provider attachments in contrasting child care settings II:...
- Ahnert, L., & Lamb, M. E. (2001). The East German child care system: associations with caretaking and caretaking...
- et al.
Sozialisationskonzepte für Kleinkinder. Wirkungen tradierter Erziehungsvorstellungen und staatssozialistischer Erziehungsdoktrinen im intra- und interkulturellen Ost-West-Vergleich [Socialization conceptions for infants. Effects of traditional and state-socialist educational doctrines in intra- and intercultural East-West comparisons]
- et al.
Maternal sensitivity and attachment in East German and Russian family networks
Family and center contributions to the adjustment of infants in full-time day care
New Directions for Child Development
(1990)- et al.
Attachment and exploratory behavior of one-year-olds in a strange situation