Poster #120 - The association between parental linguistic input and child language: A meta-analysis

• In Event: 2-110 - Poster Session 07
  In Poster Session: PS 07 Section - Language, Communication

Integrative Statement

Strong early language skills have a lasting positive impact on many developmental domains, including cognitive abilities, academic achievement, and socioemotional development. As such, research has sought to identify critical variables associated with language skills in children, with the goal of designing interventions to promote strong language development. Evidence suggests that parental linguistic input, the language that parents direct towards their children, is one such factor. However, research has yielded diverse findings regarding the magnitude of the relation between parental linguistic input and child language skills. Here, we conducted a meta-analysis to synthesize these findings, with the goal of characterizing the association between linguistic input and child language.

Relevant studies were identified through several electronic databases, resulting in 43 studies ultimately meeting inclusion criteria. We defined parental linguistic input in terms of the quantity (e.g., number of words/tokens) and quality of language (including complexity e.g., mean length of utterance and diversity e.g., number of different words, of speech). While linguistic input can also encompass a variety of other aspects of language (e.g., repetitions, questions, directives), our narrower definition was used to capture a more uniform construct.

Results indicated that the association between linguistic input and children’s language was small to moderate in size (d = .43; CI: .29-.57), supporting the hypothesis that parental linguistic input positively impacts children’s language development. No differences were found when linguistic input was assessed separately as quantity (d = .37; CI: .21-.52) or quality only (d = .48; CI: .25-.72), as such, moderator analyses were conducted using the combined linguistic input dataset. Moderator analyses revealed that effect sizes increased as the percent of female children in the sample increased (k = 43; d = .02; CI: .03-.01). This finding fits with research suggesting that girls under the age of 24 months may have better language skills than boys. Effect sizes were stronger when parental linguistic input was assessed when the child was in early childhood versus toddlerhood (k = 43; d = .01; CI: .00-.02). This may reflect an increased capacity on the part of the child to understand and benefit from linguistic input as their language capacity increases. Finally, effect sizes were larger when parental linguistic input was assessed during naturalistic observation (k = 6; d = .69; CI: .34-1.03), compared to free play (k = 16; d = .30; CI: .16-.43). It has been suggested that more naturalistic contexts yield better quality language from both parents and children, as they are less restricted in the content of their discussion. As such, future research may benefit from examining parents and children in their natural day to day routines.

Together, these findings suggest that parental linguistic input is an important factor in promoting child language development, and that the strength of this association varies based on certain characteristics of the child. This information can help to inform interventions seeking to improve child language outcomes.

Authors

• Nina Anderson, nanders@ucalgary.ca
  Presenting Author

• Susan Graham, University of Calgary
  Non-Presenting Author

• Heather Prime, University of Toronto
  Non-Presenting Author

• Jennifer Jenkins, University of Toronto
  Non-Presenting Author

• Sheri Madigan, University of Calgary and the Alberta Children’s Hospital Research Institute, Calgary
  Non-Presenting Author