The Predictive Value of Children's Understanding of Indeterminacy and Confounding for Later Mastery of the Control-of-Variables Strategy

Abstract

Prior research has identified age 9–11 as a critical period for the development of the control-of-variables strategy (CVS). We examine the stability of interindividual differences in children's CVS skills with regard to their precursor skills during this critical developmental period. To this end, we relate two precursor skills of CVS at age 9 to four skills constituting fully developed CVS more than 2 years later, controlling for children's more general cognitive development. Note that N = 170 second- to fourth-graders worked on multiple choice-assessments of their understanding of indeterminacy of evidence and of confounding. We find relations between these two precursor skills and children's CVS skills 2 years later at age 11 in planning, identifying, and interpreting controlled experiments, and in recognizing the inconclusiveness of confounded comparisons (understanding). In accordance with the perspective that both indeterminacy and confounding constitute critical, related yet distinct elements of CVS, both precursor skills contribute to the prediction of later CVS. Together, the two precursor skills can explain 39% of students' later CVS mastery. Overall, the understanding of indeterminacy is a stronger predictor of fully developed CVS than that of confounding. The understanding of confounding, however, is a better predictor of the more difficult CVS sub-skills of understanding the inconclusiveness of confounded comparisons, and of planning a correctly controlled experiment. Importantly, both precursor skills maintain interactive predictive strength when controlling for children's general cognitive abilities and reading comprehension, showing that the developmental dynamics of CVS and its precursor skills cannot be fully ascribed to general cognitive development. We discuss implications of these findings for theories about the development of CVS and broader scientific reasoning.

The complex associations between scientific reasoning and advanced theory of mind

“…Early scientific-reasoning abilities already differ substantially among children in kindergarten (Koerber & Osterhaus, 2019). Such differences in both kindergarten and elementary school have been associated with children's general cognitive abilities, including intelligence (e.g., Koerber, Mayer, et al, 2015;Peteranderl & Edelsbrunner, 2020), executive functions (e.g., Osterhaus et al, 2017;van der Graaf et al, 2016), and language skills (e.g., van de Sande et al, 2019;van der Graaf et al, 2016). Individual differences are also affected by family variables, particularly the parental education level (Koerber, Mayer, et al, 2015).…”

The complex associations between children’s scientific reasoning and advanced theory of mind

“…Early scientific-reasoning abilities already differ substantially between children in kindergarten (Koerber & Osterhaus, 2019). Such differences in both kindergarten and elementary school have been associated with children's general cognitive abilities, including intelligence (e.g., Koerber, Mayer, et al, 2015;Peteranderl & Edelsbrunner, 2020), executive functions (e.g., Osterhaus et al, 2017;van der Graaf et al, 2016), and language skills (e.g., van de Sande et al, 2019;van der Graaf et al, 2016). Individual differences are also affected by family variables, particularly the parental education level (Koerber, Mayer, et al, 2015).…”

The personal epistemology of parents predicts the development of scientific reasoning in children aged 6–10 years

“…Attributing knowledge to third agents and evidence talk are just two aspects of the comprehensive scientific reasoning skills that develop during kindergarten and elementary school (Koerber et al., 2015; Koerber & Osterhaus, 2019). Individual differences in these comprehensive scientific reasoning skills have been linked to several aspects of the general cognitive development of children, including their intelligence (Koerber et al., 2015; 2017; Mayer et al., 2014; Osterhaus et al., 2017; Peteranderl & Edelsbrunner, 2020), language abilities (Koerber et al., 2015; Mayer et al., 2014; Osterhaus et al., 2017; Peteranderl & Edelsbrunner, 2020; van de Sande et al., 2019; van der Graaf et al., 2016), and executive functions (Osterhaus et al., 2017; van der Graaf et al., 2016). Alongside these child‐based variables, science achievement gaps, which emerge early on and persist over time (Kähler et al., 2020; Morgan et al., 2016), have been associated with family affluence, with elementary‐school children from families with higher parental education levels performing better in many scientific reasoning tasks than those from families with lower parental education levels (Koerber et al., 2015).…”