Everett Mahaffy
Tracy M. Zehner
Lauren Westerberg3
David James Purpura
Sara A. Schmitt- 1Department of Special Education and Clinical Sciences, College of Education, University of Oregon, Eugene, OR, United States
- 2Department of Counseling Psychology and Human Services, University of Oregon, Eugene, OR, United States
- 3Wisconsin Center for Education Research, University of Wisconsin–Madison, Madison, WI, United States
- 4Department of Human Development and Family Science, Purdue University, West Lafayette, IN, United States
Children's early science knowledge is predictive of later academic outcomes, but little work has characterized the factors that shape early science development. In light of theoretical and empirical evidence highlighting the importance of social processes in science learning, the aim of this study was to examine concurrent and longitudinal relations between children's social skills and science core knowledge across 1 year of preschool. The sample consisted of 124 preschool-aged children (43% female; 70% non-white) from families at or below 127% of the federal poverty level. Using linear regression models, we examined whether teacher-rated social skills in the fall of preschool predicted science core knowledge concurrently (in the fall) and longitudinally (in the spring). After adjusting for child demographic factors, executive function, and vocabulary knowledge, no significant associations between children's social skills and science core knowledge emerged. As the first study to investigate this topic, the present study contributes to an emerging literature base. Implications for future research, including generalizability and measure selection, are discussed.
Introduction
Policymakers have identified science career preparation as a national priority, underscoring the importance of high-quality K-12 science education (Bustamante et al., 2018; National Academies of Sciences, 2021). Children from low socioeconomic status (SES) backgrounds face lower achievement in science (Betancur et al., 2018; Morgan et al., 2016). Recent research has shown that these gaps in science knowledge emerge before children begin formal schooling, such that children from low SES backgrounds enter kindergarten less prepared for science learning (Morgan et al., 2024). Considering that children's early science skills are highly predictive of their achievement in later grades (Grissmer et al., 2010; Morgan et al., 2016, 2024), these early-emerging and persisting gaps underscore the importance of identifying factors associated with children's early science knowledge, particularly for children from low SES backgrounds. However, early childhood science remains a relatively understudied area (Bustamante et al., 2017; Greenfield et al., 2009), hindering progress toward understanding factors driving these disparities. Among these potential factors, social-emotional competence (SEC) plays a key role in learning and behavior (Denham and Brown, 2010), and has been linked to early academic outcomes such as numeracy (Zehner et al., 2024). However, little work has examined the role of SEC in supporting children's science development. The present study addresses this gap by examining whether SEC predicts science core knowledge in early childhood.
Children's early knowledge of scientific phenomena is predictive not only of future science outcomes (Morgan et al., 2016), but also of later reading and math achievement (Grissmer et al., 2010). Thus, science core knowledge can be considered a key aspect of school readiness (Greenfield et al., 2009). Although traditional (e.g., Piagetian) theories of cognitive development initially posited that young children were not yet capable of scientific reasoning (for a review, see Koerber and Osterhaus, 2019), more recent findings have demonstrated that young children can reason scientifically and understand scientific phenomena when presented with developmentally appropriate ways to do so (Fragkiadaki et al., 2023; Koerber and Osterhaus, 2021; Mantzicopoulos et al., 2009; Piekny and Maehler, 2013). This is indeed reflected in preschool learning standards, with a recent content analysis finding that most state standards require young children to understand concrete and abstract science facts such as attending to patterns unfolding over time (Ocasio et al., 2021).
Children's science development has been conceptualized in a multitude of ways over time (Allen and Kambouri-Danos, 2017). Such conceptualizations frequently separate at least two domains of science understanding, such as knowledge of science facts and broader understandings of the scientific inquiry process (Allen and Kambouri-Danos, 2017; Piekny and Maehler, 2013; Tolmie et al., 2016). The Next Generation Science Standards (NGSS; NGSS Lead States, 2013) further separate science learning into three domains: disciplinary core ideas (“facts”), science and engineering practices (“doing science”), and crosscutting concepts (“connecting science”), and these domains work in tandem. For instance, engaging in the scientific inquiry process (practices) may lead to new conceptual understandings (core ideas; O'Connor et al., 2021). To illustrate, a child may engage in question-asking and prediction-making about whether an object would float or sink followed by hypothesis-testing by placing the object in water (Meindertsma et al., 2014). Through this process, the child would then integrate their observations into their conceptual understanding of scientific ideas such as “heavy” and “light” as well as emergent understandings of density. In this way, science core knowledge can be considered a measurable outcome of the scientific inquiry process (Guo et al., 2015). Further, one study found that general knowledge of the world (i.e., science core knowledge) in first grade was the strongest predictor of children's science achievement in third to eighth grade, with kindergarten core knowledge being highly predictive of first-grade knowledge (Morgan et al., 2016). The present study, therefore, centers on science core knowledge as a key measurable outcome of the science learning process and dimension of academic readiness.
As children engage in the scientific inquiry process, they draw upon a range of social, behavioral, and self-regulatory competencies. A particularly salient predictor of science outcomes in early childhood is executive function (EF; Anthony and Ogg, 2020; Gropen et al., 2011; Kim et al., 2021; Morgan et al., 2024; Nayfeld et al., 2013). Children with higher EF experience larger gains in science knowledge during preschool and kindergarten (Anthony and Ogg, 2020; Koerber and Osterhaus, 2021) and some evidence indicates that EF may even mediate the relation between low SES and lower science performance (Morgan et al., 2024). However, these findings are not ubiquitous across studies; for example, one study found no concurrent associations between EF and science core knowledge (Westerberg et al., 2021). Regarding the mechanism linking EF and science outcomes, one study found that the relation between EF and science learning was mediated by learning-related behaviors such as attending to instruction and persevering through difficult tasks (Anthony and Ogg, 2020). Thus, previous research has broadly implicated self-regulatory skills in early science learning, which, in turn, support children's ability to engage in specific behaviors that promote learning. However, the majority of work to date has investigated only cognitive and behavioral aspects of self-regulation rather than social-emotional regulatory processes. Indeed, to our best knowledge, no studies to date have characterized the relation between SEC and science outcomes, presenting a critical gap in the literature.
Young children are often considered “natural scientists” due to their innate curiosity about the world (Allen and Kambouri-Danos, 2017; Gopnik, 2012). In this way, children's initial conceptions of scientific phenomena are formed through direct experience with the world around them (Allen and Kambouri-Danos, 2017; Fragkiadaki et al., 2023). Adults, including caregivers and educators, play a central role in this process by answering questions and engaging in conversations about scientific phenomena (Bae et al., 2023; Bustamante et al., 2023; Junge et al., 2021; O'Connor et al., 2021; Westerberg et al., 2022). Thus, science learning in early childhood is an inherently social process (Fleer, 2013). This remains true once children enter preschool, as science activities often take place in group settings (Bustamante et al., 2018; Tolmie et al., 2016). Similarly, play-based science learning has been shown to support children's development of scientific conceptual understanding (Adbo and Vidal Carulla, 2020; Guarrella et al., 2023).
Given the inherently social nature of early science learning, children's SEC may play a key role in supporting science development in both home and school environments. Observational studies have found that preschool children demonstrating successful science problem-solving skills often rely on collaboration and support-seeking, highlighting the social nature of science problem solving in early childhood (Fusaro and Smith, 2018). It holds, then, that children with stronger SEC may be better able to engage in these processes such as discussion and collaboration, frequent features of group- and play-based science learning activities. Within SEC, social skills encompass a range of competencies which may be particularly important in promoting children's ability to engage in science learning activities. For example, such skills include communication, cooperation, responsibility, and engagement (Gresham and Elliott, 2008). Children with these foundational social skills are likely better able to (1) engage in social classroom learning activities such as play-based and collaborative tasks, and (2) engage in effective inquiry-based conversations with teachers, caregivers, and peers (i.e., “sustained shared thinking”; Adbo and Vidal Carulla, 2020), in turn supporting science core knowledge development. The present study, therefore, focuses specifically on social skills as a domain of SEC particularly relevant to early science learning.
Considering the pervasive and early-emerging science achievement gaps, early childhood may be a particularly impactful window for science-focused interventions (Bustamante et al., 2017; Greenfield et al., 2009). However, existing interventions delivered in preschool settings have demonstrated mixed effectiveness across studies and populations (Greenfield et al., 2009). This inconsistent effectiveness may indicate that child-level characteristics such as SEC are especially crucial determinants of science achievement disparities (Morgan et al., 2024) beyond exposure to science content alone. Thus, elucidating these characteristics that support science learning is crucial for developing more effective interventions. For instance, the present study has potential to inform future work by clarifying the role of SEC in science outcomes, which may be used to develop integrated interventions that promote SEC in addition to targeted science instruction, optimizing intervention efficacy (Garner et al., 2018; Gropen et al., 2011).
The present study
Given the social nature of science learning in home and classroom settings (Bustamante et al., 2018; Tolmie et al., 2016), the present study aims to quantify the concurrent and longitudinal relations between social skills (a component of SEC) and science core knowledge in early childhood. We chose to examine both concurrent and longitudinal relations, as we suspected that social skills would be particularly important for children's gains in science core knowledge across 1 year of preschool, as skills such as cooperation may promote or hinder children's ability to engage in collaborative science learning in the classroom. The two preregistered research questions that guided our analyses are as follows: (1) Do social skills concurrently predict science core knowledge in the fall of preschool when adjusting for child demographic factors, EF, and vocabulary? (2) Do social skills in the fall of preschool predict science core knowledge in the spring of preschool when adjusting for child demographic factors, EF, and vocabulary knowledge? Although research on social skills and science in early childhood is limited, our hypotheses were informed by previous research demonstrating a link between SEC and early math (Dobbs et al., 2006; Doctoroff et al., 2016; Mackintosh and Rowe, 2021; Zehner et al., 2024) and observational evidence that children's successful science problem solving is typically collaborative in nature, implicating social skills such as communication and cooperation in science learning (Fusaro and Smith, 2018). We hypothesized that higher social skills in fall of preschool will be associated with (1) higher scores on a science core knowledge assessment in the fall of preschool, and (2) higher science core knowledge as assessed in the spring of preschool, after adjusting for child demographic factors and baseline science core knowledge.
Method
Procedure
The present study draws upon data from a longitudinal, quasi-experimental evaluation of a state-developed preschool program (N = 126). Families were recruited from early childhood education providers, some of which qualified as state-designated high-quality programs while others served as a quasi-experimental comparison condition. The study team was provided with lists of approved providers according to the state quality rating system and a list of lower-rated childcare providers. Recruitment began with approved providers in the largest nearby counties; all eligible providers in these counties were invited to participate. For the comparison group, providers in the same counties as the state-approved programs selected for participation were contacted first, with later recruitment to additional neighboring counties until the target sample size was reached. Children were assessed in the fall and spring of preschool. Trained research assistants conducted assessments across 2–4 sessions that lasted approximately 20–30 min. In addition, preschool classroom teachers reported on children's SEC by completing the teacher-informant version of the Social Skills Improvement System (SSIS) questionnaire, which assesses SEC across two domains: social skills and problem behaviors. Questionnaires were distributed in hard-copy form during data collection and retrieved at the end of the semester. Teachers and parents were compensated $20 for each timepoint (i.e., fall and spring) that they completed surveys.
Participants
Participants were preschool-aged children in a Midwestern U.S. state who were required to be 4 years of age by August first of the school year. Children were eligible if their household income fell at or below a threshold of 127% of the federal poverty level. Due to substantial missing data, two participants were removed for analyses (see Analytic Approach). The resulting analytic sample (N = 124) consisted of 78 children participating in state-designated high-quality preschool programs and 46 enrolled in comparison programs. The sample was reasonably balanced in terms of child sex (43% female). Among the analytic sample, the majority (70%) of children identified as a race other than white. The mean monthly household income was $1,687.61 (SD = 959.83) with a range from $0 to $6,257.77.
Measures
Social-emotional competence: Social Skills Improvement System (SSIS)
The SSIS assesses children's SEC across two domains: Social Skills and Problem Behaviors (Gresham and Elliott, 2008). The Social Skills scale consists of 46 items in seven subdomains: communication, cooperation, assertion, responsibility, empathy, engagement, and self-control. The Problem Behaviors scale was not analyzed here due to the study's focus on social skills as a promotive factor in children's science development. Items are rated on a 4-point scale: 0 (never), 1 (seldom), 2 (often), and 3 (almost always). Internal consistency for this sample was α = 0.91 for the Social Skills domain. The present analysis used standard scores, using age-based norms, for the Social Skills composite.
Science core knowledge: Circle Progress Monitoring System: Science and Engineering Subtest (Circle)
The Circle is a direct assessment of young children's science knowledge across multiple domains, including physical science, life science, earth and space science, and engineering and technology (Zucker et al., 2016). The assessment consists of 24 multiple-choice items. For each item, children are shown three pictures and asked to point to the one that best answers a science-related question. All items are administered, and children receive one point for each correct response, resulting in a total score ranging from 0 to 24. In this sample, Cronbach's alpha values fell at α = 0.71 for Time 1 and α = 0.67 for Time 2. The present analysis used total raw scores.
Covariate measures
Based on previous research and theoretical relevance, we adjusted for the following covariates: child age in the fall of preschool, sex [0 = male, 1 = female], race/ethnicity [0 = white, 1 = non-white], monthly family income, group assignment [0 = comparison, 1 = state-designated high-quality (HQ)], baseline receptive vocabulary [Peabody Picture Vocabulary Test IV (PPVT)], and baseline EF [Dimensional Change Card Sort (DCCS)]. Child age was selected to account for developmental differences, whereas sex was selected to account for differences in socialization processes and social-emotional-behavioral outcomes (Bando et al., 2024). We also adjusted for child race and ethnicity as well as family income given demonstrated disparities in science outcomes across racial groups and income levels (Bali and Alvarez, 2004; Morgan et al., 2016, 2024). Finally, adjustment for baseline vocabulary and EF—which have been frequently linked to science outcomes (Anthony and Ogg, 2020; Gropen et al., 2011; Kim et al., 2021; Morgan et al., 2024; Nayfeld et al., 2013; Westerberg et al., 2021)—allowed for better isolation of the unique relation between social skills and science core knowledge.
Analytic approach
All analyses were conducted in R (R Core Team, 2025). After inspecting missingness patterns, two participants were found to be missing all three primary predictor/outcome variables of interest: SSIS in the fall, Circle in the fall, and Circle in the spring. These participants were removed prior to addressing missing data using multiple imputation, resulting in a final analytic sample of 124 participants. Little's MCAR test (Little, 1988) indicated that data were missing completely at random (p = 0.30). Missing data were handled using multiple imputation using the mice package (van Buuren and Groothuis-Oudshoorn, 2011; CART method) with 20 imputed datasets and 10 iterations. To determine the extent to which observations were nested within preschool classrooms or school sites, intraclass correlation coefficients (ICCs) were calculated for fall and spring scores. ICCs were calculated using both non-imputed and imputed datasets as a sensitivity check. Classroom-level clustering was observed to be extremely low across both timepoints. For preschool site, ICCs were near-zero for imputed and non-imputed models using fall Circle scores. For spring Circle scores, non-imputed models resulted in an ICC of 0.033, while imputed models resulted in a near-zero ICC. In accordance with our preregistered analytic plan, we then selected multilevel models including a random intercept statement for preschool center for RQ2. However, multilevel models resulted in singular fit errors. Thus, for both research questions, we instead estimated ordinary least squares linear regression models using Huber-White Sandwich cluster-robust standard errors (Robitzsch et al., 2024). Specifically, models predicted science core knowledge in the fall (RQ1) and spring (RQ2) from baseline social skills in the fall of preschool. All models adjusted for the covariates above. Longitudinal analyses included baseline Circle scores as an additional covariate to examine residualized gains over time. An ANOVA was used to compare model fit between unadjusted and adjusted models. All results were evaluated at a threshold of α = 0.05 as preregistered.
Results
Descriptive statistics
Descriptive statistics are presented in Table 1, including means, standard deviations, and ranges for primary predictor, outcome, and covariate variables. Descriptives are presented for the entire analytic sample (N = 124) prior to imputation. The average SSIS Social Skills standard score was 94.13 (SD = 15.83), falling within the average range according to age-based norms (Gresham and Elliott, 2008). Children were rated within a wide range on this measure (49–122) indicating substantial variability in the sample. The average Circle score in the fall of preschool was 16.36 (SD = 3.79) and the average Circle score in the spring was 18.35 (SD = 3.21). For covariate measures, the mean PPVT score was 92.52 (SD = 13.90) while the average DCCS score was 10.31 (SD = 5.36). Missing data were relatively minimal with the exception of fall Circle measure, with 31% of participants missing this measure. Bivariate correlations for all variables of interest are reported in Table 2; these were calculated based on the analytic sample prior to imputation.
Table 1. Descriptive statistics for predictor, outcome, and covariate variables.
Table 2. Bivariate correlations for predictor, outcome, and covariate variables.
Research question 1
Results from unadjusted and covariate-adjusted models are reported in Tables 3 and 4, respectively. In both models, teacher-reported social skills were not significantly associated with concurrent scores on the Circle science core knowledge assessment (β = 0.032, p = 0.224, CI = [−0.019, 0.083] for unadjusted models and β = 0.001, p = 0.957, CI = [−0.040, 0.042] for covariate-adjusted models). Results from an ANOVA model comparison indicated that the covariate-adjusted model demonstrated significantly better model fit (p < 0.001). Model assumptions were assessed using diagnostics for linearity, heteroscedasticity, outliers, and normality of residuals. Slight non-normality of residuals was detected at the upper tail for both models; however, upon inspecting diagnostic plots, this was determined to be minimal and assumptions were therefore reasonably well-met. No outliers were detected based on model diagnostics. Given the presence of multiple very low observations (≥3 standard deviations), models were also run without these participants as a sensitivity check. Removing these participants did not substantively change the results, therefore, findings reported here utilize the entire analytic sample.
Table 3. Model summary results for unadjusted linear regression model predicting concurrent science core knowledge from social skills.
Table 4. Model summary results for covariate-adjusted linear regression model predicting concurrent science core knowledge from social skills.
Research question 2
Results from unadjusted and covariate-adjusted models are reported in Tables 5 and 6, respectively. When adjusting only for baseline science core knowledge, initial teacher-reported social skills were significantly associated with gains on the Circle science core knowledge assessment from fall to spring (β = 0.032, p = 0.048, CI = [0.000, 0.064]). However, in the model adjusting for all covariates (age, sex, race, family income, vocabulary, and EF), baseline teacher-reported social skills no longer remained significantly associated with gains on the Circle science core knowledge assessment from fall to spring (β = 0.026, p = 0.090, CI = [−0.004, 0.055]). Results from an ANOVA model comparison indicated that the covariate-adjusted model demonstrated significantly better model fit (p < 0.001). Model assumptions were assessed as above. Some heteroscedasticity was detected in these models, which was addressed through the use of Huber-White sandwich standard errors. Slight non-normality of residuals was also detected at the lower tail for both models, however, upon inspecting diagnostic plots, this was minimal and assumptions were therefore reasonably well-met.
Table 5. Model summary results for unadjusted linear regression model predicting longitudinal science core knowledge from baseline social skills.
Table 6. Model summary results for covariate-adjusted linear regression model predicting longitudinal science core knowledge from baseline social skills.
Discussion
This study investigated whether teacher-reported social skills were significantly associated with preschool children's science core knowledge assessment scores concurrently and longitudinally across the preschool year. In contrast with our hypotheses, social skills were not a significant predictor of concurrent or longitudinal science core knowledge after accounting for vocabulary and EF in the fall of preschool, and demographic factors. To our knowledge, the present study is the first to investigate social skills as a predictor of science outcomes in preschool children. These results, although unexpected, therefore provide a novel contribution to a growing literature examining factors that influence young children's science development.
We hypothesized that social skills would be a significant predictor of science core knowledge in both concurrent and longitudinal analyses. This hypothesis was rooted in theories of early science development, which emphasize the role of social interactions in shaping children's early scientific understandings of the world (Allen and Kambouri-Danos, 2017; Fleer, 2013; Fragkiadaki et al., 2023). Additionally, an adjacent body of research has demonstrated links between SEC and early numeracy outcomes (e.g., Dobbs et al., 2006; Doctoroff et al., 2016; Mackintosh and Rowe, 2021; Zehner et al., 2024), implicating SEC as an important factor in early academic outcomes. Previous evidence indicates that factors such as EF are predictive of both math and science, and these outcomes share similar associations when studied in tandem (Blums et al., 2017). Thus, it is surprising that social skills did not significantly predict science outcomes in this study, given past findings linking SEC and early numeracy.
Our hypothesis that social skills would predict children's gains in science core knowledge over 1 year of preschool was rooted in the idea that these skills may influence children's ability to effectively participate in group- and play-based science learning activities over the preschool year, resulting in higher science core knowledge gains for children with more robust social skills. In this way, it was particularly surprising that no longitudinal relation between social skills and science core knowledge was detected after adjusting for covariates. Previous work has indicated that the amount of time dedicated to science instruction varies widely by preschool classroom (Piasta et al., 2014), and that inquiry-based science instruction is absent in many early learning settings. It is possible that some of the children in this study were not exposed to the type of inquiry-based, collaborative science learning activities that would draw directly upon children's social skills. Previous work in this sample has found associations between social skills–particularly cooperation—and early numeracy (Zehner et al., 2024). It may be the case, then, that children in this study were exposed to more collaborative mathematics instruction than science learning. Thus, social skills may play a larger role if science instruction is playful, social, and rich in exploration; however, this type of science learning is not ubiquitous in early childhood classrooms.
Vocabulary was the most salient predictor of both concurrent and longitudinal science outcomes in this sample (Westerberg et al., 2021). In our longitudinal models, the variance in science outcomes initially explained by social skills became non-significant once vocabulary was added to the model. Thus, it is possible that social skills may play a role in science outcomes, particularly if science instruction is playful and collaborative in nature, but that this association was overshadowed by the importance of vocabulary in this sample. This finding raises several additional questions. For instance, our sample consisted entirely of children from low-SES households. Previous research has demonstrated SES-related differences in language and vocabulary development (Hart and Risley, 2003), although these findings have not been replicated across studies (Sperry et al., 2019). Thus, in this predominantly low-SES sample, variance in vocabulary exposure may serve as an especially strong predictor of science core knowledge, but whether this remains true in the general population remains unclear. Indeed, previous research has implicated vocabulary as a particularly salient predictor of academic achievement in low-SES preschool samples specifically (Ramsook et al., 2020) and an explanatory factor in SES-related science disparities (Jang et al., 2024). Alternatively, it is possible that the measure used to assess students' science learning draws upon science skills that are more strongly related to vocabulary than social skills.
The science measure used in this study assessed children's science core knowledge specifically, one of three dimensions of science learning according to the NGSS (NGSS Lead States, 2013), which provides a useful framework for conceptualizing science readiness in early childhood settings. Although we selected science core knowledge as it has been established as a key aspect of school readiness (Greenfield et al., 2009; Grissmer et al., 2010), it is possible that the unmeasured dimensions of science learning draw more upon social skills and other dimensions of SEC. For instance, science practices, as defined by the NGSS, include skills such as “asking questions,” “exploring solutions,” and “communicating information.” These skills may map more closely onto social skills such as communication and cooperation, in contrast to science core knowledge, which may have stronger roots in language and vocabulary development. This has been noted in prior work investigating self-regulation and science core knowledge, for instance, Westerberg et al. (2021) did not find an association between EF and science core knowledge using the Circle despite previous research demonstrating this link. Thus, more research is needed to characterize the role of SEC in promoting science learning across all three domains.
Limitations and future directions
The present study contributes to the existing literature by being the first to investigate associations between SEC and science outcomes in early childhood. However, several limitations should be recognized in interpretation of the present study's results. A primary limitation of the present study is the reliance on indirect (i.e., teacher-reported) measures of social skills. Although previous studies using this measure have detected associations with academic outcomes (e.g., Zehner et al., 2024), teacher-rated indirect measures may be vulnerable to bias (Martinsone et al., 2022). Thus, future research should integrate multi-informant and direct observation measures to provide additional validity to social skills assessments. Additionally, as discussed, the measure used to assess science knowledge primarily assessed science core knowledge. Thus, it remains unclear whether SEC may be a significant predictor of other dimensions of science development (e.g., science practices).
An additional limitation of the present study is generalizability, as the sample was wholly comprised of students from low-SES backgrounds. Although investigating academic outcomes in this population is critical for addressing existing gaps in academic achievement, it limits the generalizability of these results. For instance, it is possible that SEC would significantly predict science outcomes in higher-SES samples. Additionally, the present project used data collected in a quasi-experimental evaluation of a state preschool program. Although the present study used treatment condition as a covariate, participants may have nonetheless received varying quality early childhood instruction. Finally, although well-powered for the analytic procedures used in this study, the sample size precluded the use of methods such as structural equation modeling which has been used in previous similar studies (e.g., Zehner et al., 2024) to investigate how individual domains of SEC are differentially associated with academic outcomes.
The present study serves as a foundation for future work investigating preschool science development. Future research could expand upon this work by assessing all three dimensions of science learning as defined by NGSS. However, more work is needed to establish validated measures that do so (Brenneman, 2011; Greenfield, 2015; Guarrella et al., 2023). Additionally, future studies may investigate these associations in larger and more SES-diverse samples to explore whether the findings here generalize to the general population and whether individual social skills have stronger predictive power than others.
Conclusion
This study examined social skills as a concurrent and longitudinal predictor of science core knowledge in a sample of preschool children. Our results found no significant associations between social skills and science core knowledge when accounting for EF, vocabulary, and child demographic factors. The present study is the first to investigate this association, however, more research is needed to determine whether these unexpected findings are generalizable beyond this sample.
Data availability statement
The data analyzed in this study is subject to the following licenses/restrictions. The dataset associated with this paper is not currently publicly available. Requests to access these datasets should be directed to c3NjaG1pdHRAdW9yZWdvbi5lZHU=.
Ethics statement
The studies involving humans were approved by the Purdue University Institutional Review Board (IRB). The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation in this study was provided by the participants' legal guardians/next of kin.
Author contributions
EM: Writing – review & editing, Writing – original draft. TZ: Writing – review & editing. LW: Writing – review & editing. DP: Writing – review & editing. SS: Writing – review & editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the Indiana's Family and Social Services Administration (FSSA) under Contract #F1-7915-PK-037 and Contract #0000000000000000000026332 to Purdue University.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
Adbo, K., and Vidal Carulla, C. (2020). Learning about science in preschool: play-based activities to support children's understanding of chemistry concepts. Int. J. Early Child. 52, 17–35. doi: 10.1007/s13158-020-00259-3
Allen, M., and Kambouri-Danos, M. (2017). Substantive conceptual development in preschool science: contemporary issues and future directions. Early Child Dev. Care 187, 181–191. doi: 10.1080/03004430.2016.1237561
Anthony, C. J., and Ogg, J. (2020). Executive function, learning-related behaviors, and science growth from kindergarten to fourth grade. J. Educ. Psychol. 112, 1563–1581. doi: 10.1037/edu0000447
Bae, J., Shavlik, M., Shatrowsky, C. E., Haden, C. A., and Booth, A. E. (2023). Predicting grade school scientific literacy from aspects of the early home science environment. Front. Psychol. 14:1113196. doi: 10.3389/fpsyg.2023.1113196
Bali, V. A., and Alvarez, R. M. (2004). The race gap in student achievement scores: longitudinal evidence from a racially diverse school district. Policy Stud. J. 32, 393–415. doi: 10.1111/j.1541-0072.2004.00072.x
Bando, R., Lopez-Boo, F., Fernald, L., Gertler, P., and Reynolds, S. (2024). Gender differences in early child development: evidence from large-scale studies of very young children in nine countries. J. Econ. Race Pol. 7, 82–92. doi: 10.1007/s41996-023-00131-1
Betancur, L., Votruba-Drzal, E., and Schunn, C. (2018). Socioeconomic gaps in science achievement. Int. J. STEM Educ. 5:38. doi: 10.1186/s40594-018-0132-5
Blums, A., Belsky, J., Grimm, K., and Chen, Z. (2017). Building links between early socioeconomic status, cognitive ability, and math and science achievement. J. Cogn. Dev. 18, 16–40. doi: 10.1080/15248372.2016.1228652
Brenneman, K. (2011). Assessment for preschool science learning and learning environments. Early Child. Res. Pract. 13:n1.
Bustamante, A. S., Bermudez, V. N., Ochoa, K. D., Belgrave, A. B., and Vandell, D. L. (2023). Quality of early childcare and education predicts high school STEM achievement for students from low-income backgrounds. Dev. Psychol. 59, 1440–1451. doi: 10.1037/dev0001546
Bustamante, A. S., Greenfield, D. B., and Nayfeld, I. (2018). Early childhood science and engineering: engaging platforms for fostering domain-general learning skills. Educ. Sci. 8:144. doi: 10.3390/educsci8030144
Bustamante, A. S., White, L. J., and Greenfield, D. B. (2017). Approaches to learning and school readiness in Head Start: applications to preschool science. Learn. Individ. Differ. 56, 112–118. doi: 10.1016/j.lindif.2016.10.012
Denham, S. A., and Brown, C. (2010). “Plays nice with others”: social-emotional learning and academic success. Early Educ. Dev. 21, 652–680. doi: 10.1080/10409289.2010.497450
Dobbs, J., Doctoroff, G. L., Fisher, P. H., and Arnold, D. H. (2006). The association between preschool children's socio-emotional functioning and their mathematical skills. J. Appl. Dev. Psychol. 27, 97–108. doi: 10.1016/j.appdev.2005.12.008
Doctoroff, G. L., Fisher, P. H., Burrows, B. M., and Edman, M. T. (2016). Preschool children's interest, social-emotional skills, and emergent mathematics skills: preschool children's interest. Psychol. Sch. 53, 390–403. doi: 10.1002/pits.21912
Fleer, M. (2013). Affective imagination in science education: determining the emotional nature of scientific and technological learning of young children. Res. Sci. Educ. 43, 2085–2106. doi: 10.1007/s11165-012-9344-8
Fragkiadaki, G., Fleer, M., and Rai, P. (2023). Science concept formation during infancy, toddlerhood, and early childhood: developing a scientific motive over time. Res. Sci. Educ. 53, 275–294. doi: 10.1007/s11165-022-10053-x
Fusaro, M., and Smith, M. C. (2018). Preschoolers' inquisitiveness and science-relevant problem solving. Early Child. Res. Q. 42, 119–127. doi: 10.1016/j.ecresq.2017.09.002
Garner, P. W., Gabitova, N., Gupta, A., and Wood, T. (2018). Innovations in science education: infusing social emotional principles into early STEM learning. Cult. Stud. Sci. Educ. 13, 889–903. doi: 10.1007/s11422-017-9826-0
Gopnik, A. (2012). Scientific thinking in young children: theoretical advances, empirical research, and policy implications. Science. 337, 1623–1627. doi: 10.1126/science.1223416
Greenfield, D. B. (2015). “Assessment in early childhood science education,” in Research in Early Childhood Science Education (Dordrecht: Springer Netherlands), 353–380. doi: 10.1007/978-94-017-9505-0_16
Greenfield, D. B., Jirout, J., Dominguez, X., Greenberg, A., Maier, M., and Fuccillo, J. (2009). Science in the preschool classroom: a programmatic research agenda to improve science readiness. Early Educ. Dev. 20, 238–264. doi: 10.1080/10409280802595441
Gresham, F. M., and Elliott, S. N. (2008). Social Skills Improvement System Rating Scales: Manual. Minneapolis, MN: NCS Pearson. doi: 10.1037/t02650-000
Grissmer, D., Grimm, K. J., Aiyer, S. M., Murrah, W. M., and Steele, J. S. (2010). Fine motor skills and early comprehension of the world: two new school readiness indicators. Dev. Psychol. 46, 1008–1017. doi: 10.1037/a0020104
Gropen, J., Clark-Chiarelli, N., Hoisington, C., and Ehrlich, S. B. (2011). The importance of executive function in early science education. Child Dev. Perspect. 5, 298–304. doi: 10.1111/j.1750-8606.2011.00201.x
Guarrella, C., van Driel, J., and Cohrssen, C. (2023). Toward assessment for playful learning in early childhood: influences on teachers' science assessment practices. J. Res. Sci. Teach. 60, 608–642. doi: 10.1002/tea.21811
Guo, Y., Piasta, S. B., and Bowles, R. P. (2015). Exploring preschool children's science content knowledge. Early Educ. Dev. 26, 125–146. doi: 10.1080/10409289.2015.968240
Hart, B., and Risley, T. R. (2003). The early catastrophe: the 30 million word gap by age 3. Am. Educ. 27, 4–9.
Jang, W., Kwon, K. -A., and Horm, D. (2024). The role of language and literacy skills in science learning from kindergarten to 5th grade: mitigating gender, racial/ethnic, and socio-economic disparities. Educ. Sci. 14:994. doi: 10.3390/educsci14090994
Junge, K., Schmerse, D., Lankes, E. -M., Carstensen, C. H., and Steffensky, M. (2021). How the home learning environment contributes to children's early science knowledge—associations with parental characteristics and science-related activities. Early Child. Res. Q. 56, 294–305. doi: 10.1016/j.ecresq.2021.04.004
Kim, M. H., Bousselot, T. E., and Ahmed, S. F. (2021). Executive functions and science achievement during the five-to-seven-year shift. Dev. Psychol. 57:2119. doi: 10.1037/dev0001261
Koerber, S., and Osterhaus, C. (2019). Individual differences in early scientific thinking: assessment, cognitive influences, and their relevance for science learning. J. Cogn. Dev. 20, 510–533. doi: 10.1080/15248372.2019.1620232
Koerber, S., and Osterhaus, C. (2021). Science competencies in kindergarten: a prospective study in the last year of kindergarten. Unterrichtswissenschaft 49, 117–136. doi: 10.1007/s42010-020-00093-5
Little, R. J. (1988). A test of missing completely at random for multivariate data with missing values. J. Am. Stat. Assoc. 83, 1198–1202. doi: 10.1080/01621459.1988.10478722
Mackintosh, B. B., and Rowe, M. (2021). Baseline inequalities: social skills at preschool entry moderate math development. J. Res. Child. Educ. 35, 1–21. doi: 10.1080/02568543.2020.1728446
Mantzicopoulos, P., Samarapungavan, A., and Patrick, H. (2009). “We learn how to predict and be a scientist”: early science experiences and kindergarten children's social meanings about science. Cogn. Instr. 27, 312–369. doi: 10.1080/07370000903221726
Martinsone, B., Supe, I., Stokenberga, I., Damberga, I., Cefai, C., Camilleri, L., et al. (2022). Social emotional competence, learning outcomes, emotional and behavioral difficulties of preschool children: parent and teacher evaluations. Front. Psychol. 12:760782. doi: 10.3389/fpsyg.2021.760782
Meindertsma, H. B., van Dijk, M. W. G., Steenbeek, H. W., and Van Geert, P. L. (2014). Assessment of preschooler's scientific reasoning in adult-child interactions: what is the optimal context? Res. Sci. Educ. 44, 215–237. doi: 10.1007/s11165-013-9380-z
Morgan, P. L., Farkas, G., Hillemeier, M. M., and Maczuga, S. (2016). Science achievement gaps begin very early, persist, and are largely explained by modifiable factors. Educ. Res. 45, 18–35. doi: 10.3102/0013189X16633182
Morgan, P. L., Farkas, G., Oh, Y., and Hillemeier, M. M. (2024). Executive functions, oral vocabularies, and early literacy skills mediate sociodemographic gaps in mathematics and science achievement during elementary school. Learn. Individ. Differ. 112:102447. doi: 10.1016/j.lindif.2024.102447
National Academies of Sciences Engineering, and Medicine. (2021). Call to Action for Science Education: Building Opportunity for the Future. Washington, DC: The National Academies Press.
Nayfeld, I., Fuccillo, J., and Greenfield, D. B. (2013). Executive functions in early learning: Extending the relationship between executive functions and school readiness to science. Learn. Individ. Differ. 26, 81–88. doi: 10.1016/j.lindif.2013.04.011
NGSS Lead States (2013). Next Generation Science Standards: For States, by States. Washington, DC: National Academies Press.
Ocasio, A., Waltzer, T., Caudy, C., and Kloos, H. (2021). Are preschoolers expected to learn difficult science constructs? A content analysis of U.S. standards. J. Child. Educ. Soc. 2, 364–390. doi: 10.37291/2717638X.202123122
O'Connor, G., Fragkiadaki, G., Fleer, M., and Rai, P. (2021). Early childhood science education from 0 to 6: a literature review. Educ. Sci. 11:178. doi: 10.3390/educsci11040178
Piasta, S. B., Pelatti, C. Y., and Miller, H. L. (2014). Mathematics and science learning opportunities in preschool classrooms. Early Educ. Dev. 25, 445–468. doi: 10.1080/10409289.2013.817753
Piekny, J., and Maehler, C. (2013). Scientific reasoning in early and middle childhood: the development of domain-general evidence evaluation, experimentation, and hypothesis generation skills. Br. J. Dev. Psychol. 31, 153–179. doi: 10.1111/j.2044-835X.2012.02082.x
R Core Team (2025). R: A Language and Environment for Statistical Computing (Version 4.5.0) [Computer Software]. R Foundation for Statistical Computing. Available online at: https://www.R-project.org/ (Accessed September 2, 2025).
Ramsook, K. A., Welsh, J. A., and Bierman, K. L. (2020). What you say, and how you say it: preschoolers' growth in vocabulary and communication skills differentially predict kindergarten academic achievement and self-regulation. Soc. Dev. 29, 783–800. doi: 10.1111/sode.12425
Robitzsch, A., Grund, S., and Henke, T. (2024). Miceadds: Some Additional Multiple Imputation Functions, Especially for Mice (Version 3.17-44). Available online at: https://CRAN.R-project.org/package=miceadds (Accessed September 2, 2025).
Sperry, D. E., Sperry, L. L., and Miller, P. J. (2019). Reexamining the verbal environments of children from different socioeconomic backgrounds. Child Dev. 90, 1303–1318. doi: 10.1111/cdev.13072
Tolmie, A. K., Ghazali, Z., and Morris, S. (2016). Children's science learning: a core skills approach. Br. J. Educ. Psychol. 86, 481–497. doi: 10.1111/bjep.12119
van Buuren, S., and Groothuis-Oudshoorn, K. (2011). mice: Multivariate imputation by chained equations in R. J. Stat. Softw. 45, 1–67. doi: 10.18637/jss.v045.i03
Westerberg, L., Litkowski, E., Finders, J. K., Gerde, H. K., Duncan, R. J., Schmitt, S. A., et al. (2021). Concurrent predictors of science core knowledge in preschool. Cogn. Dev. 57:100981. doi: 10.1016/j.cogdev.2020.100981
Westerberg, L., Schmitt, S. A., Eason, S. H., and Purpura, D. J. (2022). Home science interactions and their relation to children's science core knowledge in preschool. J. Exp. Child Psychol. 222:105473. doi: 10.1016/j.jecp.2022.105473
Zehner, T. M., Paes, T. M., Devlin, B. L., Geer, E. A., Posada, G., Duncan, R. J., et al. (2024). Social-emotional competence as a predictor of early numeracy skills. Early Educ. Dev. 35, 1695–1711. doi: 10.1080/10409289.2024.2360878
Keywords: early childhood, low-income population, preschool, science, science core knowledge, social emotional competence, social skills
Citation: Mahaffy E, Zehner TM, Westerberg L, Purpura DJ and Schmitt SA (2026) Social emotional competence and science outcomes in a low-income preschool sample: a concurrent and longitudinal analysis. Front. Dev. Psychol. 4:1738899. doi: 10.3389/fdpys.2026.1738899
Received: 04 November 2025; Revised: 12 January 2026; Accepted: 14 January 2026;
Published: 10 February 2026.
Edited by:
Neil Dagnall, Manchester Metropolitan University, United KingdomReviewed by:
Ilaria Grazzani, University of Milano-Bicocca, ItalyDawn Kriebel, Immaculata University, United States
Copyright © 2026 Mahaffy, Zehner, Westerberg, Purpura and Schmitt. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Everett Mahaffy, ZXZlcmV0dG1AdW9yZWdvbi5lZHU=; Sara A. Schmitt, c3NjaG1pdHRAdW9yZWdvbi5lZHU=