Adults’ Responses to Children’s First Words. A Comparative Study With Different Social Groups.

• In Event: Poster Session 06
  In Poster Session: PS 06 Topic - Parenting & Parent-Child Relationships

Abstract

Several investigations have analyzed the impact of caregiver’s contingent responses on early language development. Contingency in adult responses predict the size of child vocabulary (Tamis-LeMonda et al., 2014; Donnellan et al., 2019), as well as the time in which the main milestones in linguistic development take place (Reed, Hirsh-Pasek & Golinkoff, 2016; Tamis-LeMonda, Bornstein & Baumwell, 2001). The majority of the studies have analyzed in US or Western European population brief sequences of caregiver- child interaction in the home or in the lab. However, in the ebb and flow of everyday-life children are embedded in interactions with multiple people in the context of diverse activities. Hence, in this study we examine the interactions in which Spanish speaking Argentinian toddlers from diverse socioeconomic backgrounds are embedded in their naturalistic household environment. Particularly we ask about the frequency of different types of responses to toddlers’ early words provided by the variety of adults and children interacting with them. Participants, 32 toddlers (9 to 20 m, 17 low SES), were audio-recorded for 4 hours at-home, without the researcher’s presence. Transcriptions were done in CHAT format. CDS was coded according to type of activity and each response to a toddler’s word utterance was identified and coded according to a system qualitatively derive from data (Strauss & Corbin, 1991). Quantitative descriptive analysis and logistic regression analysis were carried out. Results showed: a) a variety of contingent responses: repetitions and extensions of children’s utterances, requests of attention, expression of interest (¿qué? / what?, ¿eh? / ¿uh?); non-contingent responses; i.g. responses that are not related to the child utterance -imperatives and other types of directives-, and non-responses; b) SES significant differences in the number of contingent responses: middle SES children were addressed with significantly more contingent responses than low SES children, c) A higher probability of being addressed with a contingent response in middle-income households (B=0.298, SE=0.054, p<.0001) and at older ages (B=0.12, SE=0.024, p<.0001). When the activity is introduced in the regression model, SES continue to be a significant predictor but age not. When each specific type of contingent and non-contingent response is considered as dependant variable, SES and age showed different effects: extensions of children’s responses and expressions of interest are more likely in middle-SES households, while directives and non responses are more probable in low-SES families. No SES differences were observed regarding the probability of children being addressed with a repetition. Results are discussed considering studies in other populations.

Authors

• Alejandra Stein, Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)
  Presenting Author

• Celia R Rosemberg, Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)
  Non-Presenting Author

• Florencia Alam, Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)
  Non-Presenting Author

• Maia Julieta Migdalek, Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)
  Non-Presenting Author

• María Ileana Ibañez, Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)
  Non-Presenting Author