Similarities and differences in mothers’ and observers’ ratings of infant security on the Attachment Q-Sort

https://doi.org/10.1016/j.infbeh.2007.05.002Get rights and content

Abstract

This study examined the similarities and differences in maternal and observer Attachment Behavior Q-Sort ratings (AQS; Waters, 1995) and their relations to dimensions of the developmental ecology – maternal sensitivity, infant irritability, parental stress and psychosocial risk. Data was gathered from low risk (adult mothers; N = 44) and high risk mother–infant dyads (adolescent mothers; N = 83) when infants were aged 6, 10 and 15 months old, attachment being assessed at 15 months by both mothers and independent observers. A common factor was extracted from both sorts and served to operationalize the similarities between mother and observer ratings. Regressions were conducted to extract maternal and observer AQS scores that were unrelated to each other to represent the difference between the two sorts. Correlation analyses indicated that the common AQS factor was moderately linked to maternal sensitivity and parental stress, and showed a weak association to psychosocial risk and infant irritability. Residual maternal scores showed greater correlations with infant irritability and parental stress than did observer residual scores. Observer scores showed a greater correlation with psychosocial risk than maternal scores. Results suggest that common AQS variance presents a pattern of associations with ecological variables that is coherent with attachment research. Variance related to irritability, stress and risk appear to drive the differences between maternal and observer observations.

Section snippets

Mothers and observers

Since the inception of the AQS, the vast majority of studies have relied on mothers to provide infant and child security ratings. Researchers have justified the use of mothers as informants for different reasons. First, in their original report, Waters and Deane indicated mother–observer correlations ranging from .59 to .93 in different samples, with the average correlation being .80 (Waters & Deane, 1985), suggesting that mothers did not differ substantially from observers in the manner in

Maternal sensitivity

Maternal sensitivity has been shown to be an important determinant of infant attachment security and a way by which more general, ecological characteristics are able to influence infant socioemotional development in general (Belsky, 1984; De Wolff & van IJzendoorn, 1997; Pederson, Gleason, Moran, & Bento, 1998). Sensitive caregiving implies that mothers are attentive to the behaviours, emotions and other signals of their infants, and that they respond appropriately and contingently, fostering

Participants

One hundred and fourty-four mother–infant dyads participated in this study. Complete data were available for 127 dyads, currently described. Participants were recruited from two distinct subgroups: (1) low risk, adult mothers older than 20 years at the birth of the infant (n = 44); and (2) high-risk adolescent mothers, aged 20 years or less at the birth of the infant (n = 83). Mothers were recruited with the help of maternity ward nurses of two major birthing hospitals of a small (population 125 

Results

Results are presented in four sections: first, bivariate correlations are presented as an indicator of the convergence between maternal and observer AQS ratings. Correlations are also presented between each measure and each of the ecological variables currently considered. Second, a common factor analysis was conducted to isolate common variance between maternal and observer AQS ratings. Bivariate correlations were then carried out between the factor score that resulted and each of the

Discussion

The current study aimed to replicate and extend research findings regarding the association between maternal and observer assessments of attachment security using the AQS. By using a longitudinal design, involving a high and a low risk group of mother–infant dyads and assessments of different aspects of the developmental ecology, it was possible to examine whether common and divergent maternal and observer AQS variance was systematically associated with the environment that provides the context

Acknowledgement

The authors would like to thank the mothers and infants who generously donated of their time, and the following persons who provided invaluable assistance during data collection: Sophie Allaire, Caroline Bouchard, Isabelle Cabot, Julie Deslandes, Manon Fournier, Marie-France Gagnon, Isabelle Hémond, Marie Larose, Annie Lehoux, Johanne Maranda, Véronique Noreau and Julie Robitaille. This study was made possible by funding from the Fonds pour les chercheurs et l’avancement de la recherche, the

References (40)

  • J.E. Bates et al.

    Measurement of infant difficultness

    Child Development

    (1979)
  • J. Belsky

    The determinants of parenting: A process model

    Child Development

    (1984)
  • Caldwell, B. M., & Bradley, R. H. (1978). Home observation for the measurement of the environment. Unpublished manual,...
  • M.S. De Wolff et al.

    Sensitivity and attachment: A meta-analysis on parental antecedents of infant attachment

    Child Development

    (1997)
  • L. Hatcher

    A step-by-step approach to using the SAS system for factor analysis and structural equation modeling

    (1994)
  • S. Jaffee et al.

    Why are children born to teen mothers at risk for adverse outcomes in young adulthood? Results for a 20-year longitudinal study

    Development and Psychopathology

    (2001)
  • M. Main et al.

    Procedures for identifying infants as disorganized/disoriented during the Ainsworth strange situation

  • D.R. Pederson et al.

    A categorical description of attachment relationships in the home and its relation to Q-sort measures of infant-mother interaction

  • D.R. Pederson et al.

    Expressions of the attachment relationship outside for the strange situation

    Child Development

    (1996)
  • D.R. Pederson et al.

    Maternal sensitivity and the security of infant-mother attachment: A Q-sort study

    Child Development

    (1990)
  • Cited by (29)

    • Paternal mind-mindedness and child executive functioning in the kindergarten classroom

      2022, Early Childhood Research Quarterly
      Citation Excerpt :

      A substantial body of research shows that the AQS is a valid measure of child attachment (for a meta-analysis, see Cadman, Diamond & Fearon, 2018). The AQS also demonstrates good interrater reliability (.87, Tarabulsy et al., 2008). In the present study, interrater reliability was conducted for 22% of the dyads at T1 and for 19% at T3 and was found to be satisfactory: ICC = .71 and .70, respectively.

    • Contextual stress and maternal sensitivity: A meta-analytic review of stress associations with the Maternal Behavior Q-Sort in observational studies

      2018, Developmental Review
      Citation Excerpt :

      The aggregation comprised a combined maternal age range of 16–45 years. A relatively strong effect contributed primarily by Tarabulsy et al. (2008) might be partly explained by effects of adversity in their high-risk subsample of adolescent mothers. Early parenthood often correlates with low SES (Bunting & McAuley, 2004; Hoff, Laursen, & Tardif, 2002; Jaffee et al., 2001; Lemelin et al., 2006), so maternal age effects in young mothers may well be a function of associated SES demands (low education and/or income).

    • My mother is sensitive, but I am too tired to know: Infant sleep as a moderator of prospective relations between maternal sensitivity and infant outcomes

      2014, Infant Behavior and Development
      Citation Excerpt :

      Agreement between the two raters’ sorts was very good, ICC = .86. The MBQS is significantly related to other measures of maternal behavior, such as the HOME Inventory and the Ainsworth scales (see Pederson & Moran, 1995), and shows good temporal stability (Bailey, Moran, Pederson, & Bento, 2007; Tarabulsy et al., 2008). Its construct validity is demonstrated by meta-analytic data showing its excellent predictive capacity with respect to child attachment security (Van IJzendoorn, Vereijken, Bakermans-Kranenburg, & Riksen-Walraven, 2004).

    View all citing articles on Scopus
    View full text