“This Bird Can't Do It ‘Cause this Bird Doesn't Swim in Water”: Sibling Teaching During Naturalistic Home Observations in Early Childhood

Abstract

Social-constructivist models of learning highlight that cognitive development is embedded within the context of social relationships characterized by closeness and intimacy (Vygotsky, 1978). Therefore, in contrast to prior research employing semistructured paradigms, naturalistic sibling-directed teaching was examined during ongoing interactions at home. Thirty-nine middle-class, two-parent families (older sibling, M_age = 6;3; younger sibling, M_age = 4;4) were studied. Intentional sequences of sibling-directed teaching were coded for: a) frequency; b) type of knowledge taught (i.e., conceptual or procedural); c) initiation of teaching by the teacher or following a request by the learner; d) teaching strategies (e.g., direct instruction, demonstration); and e) learner responses (e.g., active involvement, rejection). Findings indicated that teachers were most likely to initiate teaching spontaneously rather than respond to direct requests by learners. Teachers were most likely to initiate teaching of procedural knowledge, while learners were most likely to request the teaching of conceptual knowledge. The type of teaching strategy employed depended on who initiated the teaching and the type of knowledge taught. The response of the learner was associated with who initiated the teaching and the teaching strategies employed. These findings reveal the nuances and sophistication of young children's attempts to teach one another naturally in their homes. Findings are discussed in light of recent empirical and theoretical work.

ACKNOWLEDGMENTS

We wish to thank the following research assistants: Jessica Kurta, Ryan Persram, and Brittany Scott.

Notes

¹Data were analyzed based on a variety of statistical techniques including repeated-measures ANOVAs and MANOVAs with a) raw frequencies, b) transformed (log10 + 1) frequency scores, as well as c) proportion scores. In each case, a total of 36 effects were tested. Out of the total, 34 significance tests revealed consistent outcomes, indicating that 94% of the results were consistent across all three sets of analyses. To account for the correlation between the dependent variables and to control for Type I error, we report MANOVAs as they are a more conservative approach. The two inconsistent results (i.e., that failed to emerge as significant in the repeated-measures ANOVAs) were the initiation by knowledge interaction with transformed scores, F(1, 38) = 1.29, p = .26, and knowledge by strategy interaction with raw frequencies, F(1.66, 63.07) = 567.64, p = .06.

Additional information

Notes on contributors

Allyson Funamoto

Allyson Funamoto is now a graduate student in the Department of Educational Psychology at the University of Alberta, Edmonton, Alberta, Canada.