Poster #34 - Science Vocabulary as a Window into Differences in Children’s Science Knowledge

• In Event: 3-159 - Poster Session 12
  In Poster Session: PS 12 Section - Cognitive Process

Integrative Statement

Achievement gaps are a well-known problem in science education (U.S. Department of Education, 2000). This pertinent issue is likely to continue: these gaps may be present in increasingly large numbers of U.S. children over time as income inequality increases (Gould, Mishel, & Shierholz, 2013). Research has suggested that science achievement gaps are present by kindergarten, persist into high school, and are largely explained by children’s science knowledge at kindergarten (Morgan et al., 2016). The current study investigates what might occur in early childhood to cause these differences in science knowledge. As the science knowledge questions used to assess children are typically vocabulary questions (Morgan et al., 2016), we predicted that science vocabulary may be driving children’s science learning.
The current study measured the following variables to determine their contribution to children’s science knowledge: children’s age, gender, family socioeconomic status, ethnicity/race, English learning status, general receptive vocabulary, and science productive vocabulary. In order to measure whether children’s science vocabulary was related to more than simple science facts, children also drew two scientists to measure their general conceptions of science. Children between ages 4 and 12 (N = 86, 56 males, Mage=84 months) were presented with three tasks in a random order: a Draw-A-Scientist task in which children were asked to draw a scientist twice, the Peabody Picture Vocabulary Test for general receptive vocabulary, and the Woodcock Johnson Test 18: Science as a standardized science knowledge test. Parents also completed a demographics questionnaire and a science vocabulary checklist, which asked them to indicate science words that they had heard their child say out loud.
The regression models, with Woodcock Johnson Science Test performance as the outcome measure, are presented in Table 1. Results of linear regression models predicting science knowledge (Woodcock Johnson performance) revealed that both PPVT and productive science vocabulary predicted science knowledge above and beyond the other factors of age and demographic variables such as household income (R2=.775, p<.001). Critically, science vocabulary was the strongest predictor. These results suggest that children’s science knowledge is highly related to their vocabulary abilities. We also found relationships between children’s science vocabulary and their drawings; for example, children with a greater number of experiment-based words in their productive vocabulary (e.g., theory, observe, cause, measure, pattern, etc.) were more likely to draw scientists conducting experiments (R=.266, p=.018). An example of the relationship between drawings and experiment-based vocabulary is illustrated in Figure 1. These results suggest that children’s science vocabulary may indeed be driving children’s science knowledge and conceptions of science, and may be key to understanding the origins of these early and persistent achievement gaps. Future research plans to investigate whether certain words are more pertinent in helping children acquire science knowledge (e.g., experiment-based words), and how vocabulary-based interventions may be developed to help close existing achievement gaps.

Authors

• Emma Lazaroff, University of Wisconsin-Madison
  Presenting Author

• Megan Kaul, University of Wisconsin-Madison
  Non-Presenting Author

• Haley Vlach, University of Wisconsin-Madison
  Non-Presenting Author