Young Children’s STEM Learning from Fictional Books with Factual Information

• In Event: Poster Session 06
  In Poster Session: PS 06 Topic - Attention, Learning, Memory

Abstract

Individual differences in early STEM learning can impact learning in formal schooling (Verdine et al., 2017). Shared-book reading is a setting in which children can gain early STEM knowledge. However, STEM books vary in their quality (i.e., amount of Support for encoding STEM facts) and style (Non-Fiction = facts presented without story contexts vs. Hybrid = facts presented within story contexts). Prior research found that children learn more factual information from Non-Fiction STEM books that are High in Support compared to Low (Miller et al., 2020). However, no known prior research investigated how young children learn STEM facts from Hybrid books in the context of shared book reading. We bridge this gap in prior research by investigating young children’s factual learning from Hybrid STEM books in the context of shared book reading. We test whether: 1) levels of Support for encoding facts in Hybrid books influences STEM learning, similarly to its effect on Non-fiction books; and 2) whether children differentially learn factual versus narrative information from Hybrid books.

Forty-three caregivers read two narrative Hybrid books to their 4- to 5-year-old children (M age = 4.75 years, range = 4.07 to 5.73, 22 girls). We adapted the Non-Fiction books used in Miller et al. (2020) by adding story elements such as characters and plots. We maintained the same STEM facts from the original books, which were either High or Low in Support. Caregivers read both books to their children. Then the experimenter asked each child open-ended recall questions about information learned in the books. Half the questions were the same Expository questions used in Miller et al. asking children about central facts in the books (e.g., “why do some animals have shells?”). The other questions were Narrative, asking children about the plot (e.g., “where did Rue go on her birthday?”).

We used a within-subjects ANOVA to test whether levels of Support (High, Low) and Question Type (Expository, Narrative) influenced children’s recall. We found main effects of Support (F1,42=16.35, p <.001) and Question Type (F1,42=22.25, p <.001), as well as a Support x Question Type interaction (F1,42= 4.52, p=.040), Figure 1. Follow-up t-tests showed that for Expository Questions, children recalled more from the High than Low Support book (t42=4.23, p <.001). However, no difference was found for Narrative Questions (t42=1.68, p =.099). These findings show that children learned more narrative than expository information from the Hybrid books. However, levels of Support within the books influence children’s factual learning. Levels of Support had similar impact on STEM learning in these Hybrid books as was found for Non-Fiction books in Miller et al. (2020). We are currently extending this work by analyzing how book quality and style influence parent-child talk during shared book reading to impact individual differences in learning from the books. Overall, this work contributes to understanding the complexity of STEM learning from shared book reading and identifies ways to optimize STEM learning prior to formal schooling by exposing children to high quality STEM books.

Authors

• Melanie H. Hanft, Emory University
  Presenting Author

• Alissa G. Miller, Emory University
  Non-Presenting Author

• Hilary E. Miller-Goldwater, Emory University
  Non-Presenting Author

• Patricia J Bauer, Emory University
  Non-Presenting Author