Activity contexts and input quality and quantity in the households of Argentinian socio-economically diverse children

RAMÍREZ, MARIA LAURA; ROSEMBERG CELIA RENATA; IBAÑEZ, MARÍA ILEANA; ALAM FLORENCIA

SHEFFIELD

Simposio; Child Language Symposium 2019; 2019

Research demonstrating the impact of the quantity and quality of input on language development has mainly focused on populations from North America and Europe. Most studies have analyzed relatively brief child-parent?s dyadic interactions in the home or lab in activities clearly delimited and often child-centered, e.g. book-reading, play and meals. Their findings indicate that input properties vary regarding the type of activity and the number and features of participants, all of which are typically defined by culture and differ between social environments. Only recently, naturalistic studies have begun to analyze input properties in the context of the activities in which children participate daily (Soderstrom & Wittebolle, 2013; Tamis-LeMonda, Custode, Kuchirko, Escobar, & Lo, 2018), and a few have paid attention to less-studied cultural and social groups (Glas & Kern, 2015; Glas, Rossi, Hamdi-Sultan & Batailler, 2018). Here we examine the effects of the type of activities on input properties in a socio-economically diverse sample of Argentinian children. Thirty children (8 to 20 months) who vary regarding socio-demographic variables- mother?s education (ranging from primary to post-graduate education) and household size (ranging from 3 to 9 people)- were audio-recorded for 4 hours. CDS measures of volume (word tokens), lexical diversity (word types and VOCD) and syntactic complexity (MLU) were calculated using CLAN (MacWhinney, 2000). CDS was coded for activity and clustered into two main categories: structured activities -those organizing adult and child?s non-verbal and verbal interactions, e.g., booksharing, regulated play and adult-child conversations-, and non-structured activities, e.g., feeding, grooming, exploratory object and physical play, household chores, conversations between adults, outings, watching TV, in which the objective and the actions do not necessarily structure the verbal interaction. A descriptive analysis of input properties in structured and non-structured activities was followed by mixed effect regression analysis to estimate the effects of the activity and socio-demographic variables, controlling child age.Findings showed: a) higher values of input measures in structured activities than in non-structured ones, particularly in quantity of word tokens and types; b) the effects of the activity and mother?s education on the number of tokens (activity: B: -316.143, SE: 81.536, p