Examining the Latent Structure of Maternal Caregiving Behaviors in (Non)Distressing Caregiving Contexts

• In Event: Poster Session 10
  In Poster Session: PS 10 Topic - Parenting & Parent-Child Relationships

Abstract

Caregiving behaviors in infancy have been characterized along several key dimensions, including sensitivity (i.e., prompt, appropriate responsiveness), intrusiveness (i.e., imposing parental agenda on child), detachment (i.e., lack of emotional connection), positive regard (i.e., warmth and affection toward child), and negative regard (i.e., physical and verbal harshness toward child). Emerging evidence regarding the latent structure indicates that mothers’ caregiving behaviors are best captured by two factors, one reflecting negative intrusiveness (i.e., negative regard and controlling) and one reflecting emotional engagement (i.e., positive regard and emotional connection). Extending these findings, the current study examined the latent structure of maternal behaviors across (non)distressing caregiving contexts using confirmatory factor analysis (CFA).

The sample included 142 mothers and their 6-month-old infants. Maternal caregiving behaviors were observed during the free play and the reunion episode of the Still Face Procedure (SFP; Tronick et al., 1978), which served as the nondistressing and distressing contexts, respectively. The free play and the SFP were videotaped and coded using scales adapted from the NICHD Study of Early Child Care (Frosch & Owen, 2016). The following variables were included in analyses: global sensitivity, intrusiveness, detachment, positive regard, and negative regard.

We used CFA to examine the fit of our data to the factor structure in both contexts. Results indicated poor model fit. Factor analysis further indicated that in addition to global sensitivity, negative regard also cross-loaded on both factors. Thus, we modified the latent caregiving structure by having both sensitivity and negative regard cross-load on two factors. Configural invariance test indicated that the number of factors and the pattern of factor structure were equivalent across (non)distressing contexts (χ2 = 0.05, p = 0.98). Measurement invariance test further suggested comparable factor loadings across the two contexts (χ2 = 17.06, p = 0.05). Findings indicated that the latent structure of maternal caregiving can be best accounted by two factors: control and emotional engagement. Importantly, the extent to which mothers exhibited control varied according to the contexts. Mothers exhibited greater levels of control in the distressing (vs. nondistressing) context, indicating that mothers’ control was heightened within this context (Figure 1). Caregiving context did not moderate mothers’ emotional engagement, suggesting this type of caregiving behavior does not vary according to context.

Most research examines caregiving in nondistressing contexts, implying a tacit assumption that mothers’ behaviors are the same across contexts. Evidence from this study suggests that this assumption is not fully supported by the data. Our findings indicated that although mothers exhibit comparable levels of emotional engagement across contexts, the extent to which they exhibit controlling behavior depends on context. Differences in mothers’ controlling behavior according to context might explain discrepancies in the literature regarding the predictive significance of such parenting behavior for children’s maladjustment (Clincy & Mills-Koonce, 2013). Thus, findings inform future research practices, suggesting researchers carefully consider the context in which parenting behaviors are observed and evaluate prior evidence according to caregiving context. Further, these findings inform parenting interventions by suggesting that consideration of context is important for reducing specific negative parenting behaviors.

Authors

• Nanxi Xu, Vanderbilt University
  Presenting Author

• Ashley M. Groh, University of Missouri - Columbia
  Non-Presenting Author