Chapter 6. Summary and conclusions

This chapter summarises the main findings for the United States and discusses them in relation to themes such as gender, socio-economic status, home learning environment and early childhood education and care.

    

Early learning in the United States

In early childhood, children learn and develop faster than at any other time in their lives. The experiences they have in their earliest years can set the trajectories for their future well-being and success. Societies interested in improving the equity of children’s outcomes should focus on what happens in these earliest years. The International Early Learning and Well-being Study (IELS) was designed to provide valid, reliable and comparable information to participating countries about children’s early learning in a range of different domains, and on how this learning relates to a host of individual characteristics and contextual factors. This chapter summarises the main IELS findings for the United States.

There is room for improvement in the early learning of young children in the United States, particularly in the cognitive subdomains

In IELS, five-year-olds in the United States had significantly lower emergent literacy and emergent numeracy scores than five-year-olds in Estonia and in England. Mean scores of US five-year-olds in the self-regulation subdomains of working memory and mental flexibility were also significantly below the averages of each of the other participating countries. In the subdomain of inhibition, however, the mean score was one-quarter of a standard deviation higher than the overall mean, similar to the mean score in Estonia and significantly higher than that of England. In terms of social-emotional skills, the picture was more mixed. Specifically, five-year-olds in the United States had similar mean scores on emotion attribution and trust as children in England and Estonia. Overall, children in the United States were less able than children in Estonia to accurately identify the feelings of characters in stories and about as accurate as children in England. Children in the United States were rated by their educators as having lower prosocial skills but also less disruptive behaviour than on average in IELS, although the deviations from the overall IELS means were modest in both cases.

Children’s skills in each of the learning domains assessed in IELS were interrelated in the United States. Emergent literacy and numeracy scores were strongly positively correlated with one another, and each was moderately to strongly positively correlated with working memory, mental flexibility and emotion identification. Correlations were weaker between each of emergent literacy and emergent numeracy and scores in inhibition (directly assessed), trust and prosocial behaviour (educator-rated). The correlations between emergent literacy and disruptive behaviour, and between emergent numeracy and disruptive behaviour, were negligible.

Conclusion 1: There is room for improvement in terms of the early learning skills of young children in the United States assessed in IELS, and in particular in the areas of emergent literacy, emergent numeracy, working memory and mental flexibility. These early cognitive skills have been shown in previous research to be among the strongest predictors of later cognitive competence (Duncan et al., 2007[1]), and to be strongly associated with a range of educational, social, health and economic outcomes throughout the life course (Kautz et al., 2014[2]).

Girls display higher levels of emergent literacy and social-emotional skills than boys, but there are no gender differences in emergent numeracy or self-regulation

In the United States, five-year-old girls had a higher mean emergent literacy score than five-year-old boys. Although significant, the gap was modest, and was similar in size and direction to the gender gaps in Estonia and England. The finding was in line with findings of many national and international assessments that show girls consistently outperforming boys in literacy or reading achievement (National Assessment of Educational Progress [NAEP], Progress in International Reading Literacy Study [PIRLS], Programme for International Student Assessment [PISA], Early Childhood Longitudinal Study [ECLS-K]). Gender differences in affective factors such as reading motivation, reading interest and reading self-concept, as well as in reading behaviour, such as the nature and extent of leisure reading, have been implicated in the gender gaps in literacy found in programmes of assessment worldwide. Results from IELS show that gender gaps in the early language skills that predict later reading competence (listening comprehension, vocabulary knowledge and phonological awareness) have already started to emerge at the time children start school, before most children would be expected to be reading independently. These gaps do not seem fully attributable to differences in parent practices with girls and boys at home. In the United States, boys and girls were equally likely to be read to frequently by their parents, to have access to children’s books at home, to attend ECEC, and to have parents who were involved at school. However, parents did report that they more likely to sing songs and nursery rhymes to girls on multiple days each week than to boys, an activity that was significantly associated with emergent literacy scores after accounting for socio-economic status and gender. It is likely that factors other than those measured in IELS contribute to the gender gap in emergent literacy skills at age five.

There was no significant difference in the mean scores of boys and girls on the IELS assessment of emergent numeracy. Nonetheless, girls were more likely than boys to be rated as having above-average mathematics development by their parents and teachers, and less likely to be rated as below average. It is possible that societal beliefs about girls having lower mathematics ability may have meant that girls who showed any numerical aptitude were more likely to be rated above average by parents and educators than boys with similar ability. Gender differences in favour of boys have been found in a number of large-scale assessments of mathematics achievement with older US children, including in NAEP, PISA and the Trends in Mathematics and Science Study (TIMSS). Given that IELS showed no gender differences at the age of five, it is possible that these differences emerge later in childhood, perhaps as a result of practices at school or of cultural beliefs about gender and mathematical ability transmitted to children by peers, parents, teachers or wider society. Given the cross-sectional designs of these large-scale assessments, it is not possible to conclude this with certainty.

At five years of age, girls in the United States had significantly higher mean scores on all measures of social-emotional skills administered in IELS. They displayed greater ability to identify emotions and take others’ perspectives, skills that were directly assessed in IELS. Girls were also more trusting and displayed more prosocial and less disruptive behaviour than boys did, as indirectly assessed by their educators. The gender gaps in mean social-emotional learning scores were somewhat smaller in the United States than in either England or Estonia.

In the United States, boys were significantly more likely than girls were to have experienced learning difficulties or to have experienced social, emotional or behavioural difficulties, according to their parents, replicating the patterns found in Estonia and England. Boys from lower socio-economic backgrounds were particularly likely to have experienced these challenges and represent a group that may be at risk for later educational and social difficulties.

Conclusion 2: Particular attention may need to be paid to the literacy development and the social- emotional learning of boys in order to bring their early outcomes in line with those of girls and/or to prevent further widening of gaps.

Conclusion 3: Around the time of school entry, girls in the United States have similar early numeracy skills to boys. Action may be needed in order to avoid gender gaps emerging in favour of boys as children progress through school.

Children’s IELS scores vary considerably by socio-economic background, but the United States is faring better at achieving equity along racial and ethnic lines

Across all three countries that participated in IELS, socio-economic status was significantly positively correlated with children’s early skills. In the United States, children from higher socio-economic status (SES) backgrounds had higher scores, on average, than children from lower backgrounds in emergent literacy, emergent numeracy, working memory, mental flexibility and inhibition. They also had higher mean scores on the social-emotional subdomains of emotion identification, prosocial behaviour and trust. The exceptions to this trend related to disruptive behaviour and emotion attribution. Low-SES and high-SES children displayed similar average levels of disruptive behaviour at school and did not differ significantly in their average emotion attribution scores. The relationships between children’s socio-economic background and many of the early skills assessed by IELS were stronger in the United States than in either of the other two countries that participated in IELS. For example, the gap in mean scores in emergent literacy between children in the top quartile and the bottom quartile of SES in the United States was approximately 60% greater than the corresponding gap in Estonia. There were significant associations between SES and children having experienced low birth weight or premature birth, learning difficulties, and social, emotional or behavioural difficulties in the United States, with children from lower SES backgrounds more likely to have experienced each of these challenges.

There were no significant differences in the United States between the mean IELS scores of White, Black and Asian children or those of two or more races on the assessments of emergent literacy, emergent numeracy, any of the three self-regulation subdomains, nor on the direct and indirect assessments of social-emotional skills.1 After accounting for socio-economic background or home language, the mean emergent literacy, emergent numeracy and mental flexibility scores of Hispanic children were not significantly different from those of White children. There were also no significant associations between children’s racial or ethnic background and whether their language (receptive or expressive) development or numeracy development were rated as above average, below average or average by their teachers or by their parents. As described in Chapter 2 of this report, assessments of the educational progress of older children in the United States (such as NAEP) have consistently shown gaps in student achievement in reading and mathematics between White and Black children and between White and Hispanic children, although the size of these gaps has been closing over time. In IELS, there were also no significant associations between children’s racial or ethnic background and their having experienced low birth weight or premature birth, learning difficulties, or social and emotional difficulties.

Conclusion 4: Around the time of school entry, there are already substantial gaps in scores in cognitive and social-emotional domains between children from low- and high-SES backgrounds in the United States.

Conclusion 5: There are very few gaps in early skills along racial or ethnic lines among five-year-olds in the United States, suggesting that these gaps emerge as children progress through school.

Five-year-olds in the United States come from linguistically diverse homes and have different linguistic repertoires

In the United States, one in five children for whom information on home language was available had at least one parent who mainly spoke a language other than English at home. These children had significantly lower emergent literacy scores, on average, than children whose parents – or their sole parent – mainly spoke English at home. When interpreting this finding, it is important to note that the IELS assessment of early literacy in the United States was only administered in English, and was not translated into any of the other languages that children might use or be exposed to in their homes or neighbourhoods. While scoring lower, on average, on what was an assessment of English vocabulary, oral comprehension of English, and knowledge of English phonemes, it is important to recognise that children with other languages at home have linguistic repertoires that are spread across more than one language. These children are likely to have proficiency in two (or more) named languages, and may also use code-mixing/code-switching or translanguaging in their interactions at home or the community. The IELS assessment of emergent literacy did not provide these children any opportunity to demonstrate these linguistic proficiencies and that should be borne in mind when interpreting these findings. Bilingual or plurilingual children’s performance in an assessment of early literacy skills in one language should not be taken as evidence that they are linguistically deficient. The best available evidence suggests that children’s home languages should be given overt support in their early childhood education and care (ECEC) settings or schools, as well as at home, if they differ from the language of instruction. Developing children’s full linguistic repertoires is likely to lead to the best outcomes for children.

Children with a home language other than English also had a significantly lower score on the emergent numeracy and mental flexibility assessments after accounting for SES, but by a smaller margin than in emergent literacy. This finding was in line with findings from NAEP, which also found smaller gaps between English-language learner (ELL) students and others in mathematics than in reading (see Chapter 2). There were no significant differences in ether working memory or inhibition scores between children whose home language was English and other children. There were also no significant differences in mean scores on emotion identification, emotion attribution or prosocial behaviour. However, children with a home language other than English had significantly lower levels of reported disruptive behaviour than children whose main home language was English.

Children from immigrant backgrounds, defined here as having both parents born outside the United States (or one parent, where only information on one parent was provided), comprised 18% of children in the study for whom information on parents’ country of origin was available. Of these, 63% had a parent who mainly spoke a language other than English at home (compared to 4% of children with a non-immigrant background). After accounting for SES and home language, there were no significant differences in the emergent literacy and emergent numeracy scores of children with and without an immigrant background. Similarly, after accounting for SES and home language, there were no differences between the mean scores of children with and without an immigrant background in any of the self-regulation subdomains. In terms of social-emotional skills, children with and without an immigrant background did not differ significantly in their mean scores on emotion identification, emotion attribution or prosocial behaviour. However, five-year-olds with an immigrant background scored significantly lower on trust than children without an immigrant background and also engaged in less disruptive behaviour than children without an immigrant background, according to their teachers, after accounting for SES. The differences were not significant after accounting for home language in addition to SES, suggesting that whether the parents mainly spoke the language of the school explained more of the variation in children’s early learning outcomes than whether or not they have an immigrant background.

Conclusion 6: Children from linguistically diverse backgrounds in the United States have unique learning strengths and challenges, both of which should be supported in their ECEC settings and early schooling.

Conclusion 7: Children with a language other than English at home and children with immigrant backgrounds do not differ significantly from other children in social-emotional skills such as emotion identification, emotion attribution and prosocial behaviour. Children with a language other than English at home, however, do display lower levels of disruptive behaviour at school than their peers.

Parents are important sources of information about their children’s learning and their activities at home are associated with their children’s early skills

Parents are important sources of information about their children. In IELS, parents were asked to evaluate their children’s development in a number of areas, relative to other five-year-olds. Very low proportions of children in the United States had parents who indicated that they were below average with respective to their receptive and expressive language development, numeracy development, self-regulation development, and social-emotional development, with over 90% of children having parents who indicated that their development as average or above average. Children’s parents were substantially less likely to describe them as below average in each of these areas than their educators were. This may reflect educators’ greater experience with children in the target age group, leading them to have a more accurate understanding of the average development of five-year-olds. Alternatively, it may reflect differences in how children display early skills in the school environment and how they do so at home, with the people who know them best. It is possible that the lack of alignment between parents’ and educators’ perceptions of children’s development may have implications for the levels of support for that learning domain given at home or at school. There may therefore be a role for increased communication between educators and parents, who are potentially important sources of information for each other on children’s learning and development.

IELS found a number of parental practices that are positively associated with children’s learning. For example, regardless of socio-economic background, five-year-olds in the United States who were read to most often by their parents had significantly higher mean scores for emergent literacy and emotion identification than children who were read to less frequently. Having access to more children’s books at home was associated with higher scores across a range of cognitive and social-emotional subdomains. Children whose parents frequently sang songs or nursery rhymes to them had better average scores on the emergent literacy assessment than children whose parents did so less often. For other activities, moderate frequency was associated with higher IELS scores than lower or higher frequency. For example, children whose parents took them to special activities (such as sport, dance or scouts) on one or two days a week had higher mean literacy scores than both those who never attended such activities and those who attended them most days.

In addition to involvement in home-based activities, parental involvement at school was positively associated with children’s scores in a range of learning domains. Children whose parents were described by teachers as being moderately or strongly involved in activities at the school had significantly higher mean scores on the assessments of emergent literacy, emergent numeracy, prosocial behaviour, trust and non-disruptive behaviour than children whose parents were not or only slightly involved, once socio-economic status was accounted for.

Children who used digital devices with moderate frequency, whether for educational activities or otherwise, had higher mean literacy scores than children who never or hardly every used them and those who used them every day, after accounting for socio-economic status. In the United States, almost half of five-year-olds used these devices every day, a higher proportion than in either England (39%) or Estonia (39%).

Conclusion 8: There is some misalignment between teachers’ and parents’ ratings of children’s development, across a range of learning domains in IELS. Increased communication between the home and the school about children’s progress is likely to be beneficial for all parties.

Conclusion 9: A range of parental practices are positively associated with children’s early learning scores in IELS. This suggests a number of ways parents might be usefully advised to support their children’s early learning and development at home.

Conclusion 10: The findings suggest that for some home activities, more is better (such as reading to a child, or singing songs and nursery rhymes), while for others, the highest mean scores are associated with moderate levels of engagement (e.g. extracurricular activities, use of digital devices).

ECEC attendance is associated with higher emergent literacy and numeracy scores, regardless of socio-economic status, but disadvantaged children are less likely to attend

In the United States, there was a statistically significant association between ECEC attendance and socio-economic status, with over 90% of five-year-olds in the top SES quartile having attended before the age of five, compared to 73% in the lowest SES quartile. There was no significant association between ECEC attendance and children’s race/ethnicity, and girls and boys were equally likely to have attended. Children in the United States who had ever attended ECEC had significantly higher mean emergent literacy and emergent numeracy scores than children who never attended, even after accounting for SES. Age of first entry to ECEC was not significantly related to scores in any of the learning outcomes assessed in IELS. ECEC participation rates were lower in the United States than in England and Estonia, where there is near-universal participation.

Conclusion 11: Regardless of socio-economic background, children in the United States who had attended ECEC have better emergent literacy and numeracy skills as assessed in IELS than those who did not. US children from higher SES households attended ECEC earlier and for longer than children from lower socio-economic backgrounds.

Conclusion 12: Expanding access to high-quality ECEC in the United States may be a means of improving early learning outcomes.

Findings from IELS in the United States show that disparities in children’s early skills related to their gender, socio-economic status and home language are already present shortly after school entry. These gaps are observable across a range of early learning domains. That some groups of children are already falling far behind others at the age of five should be cause for concern. There are groups of children in the United States who are likely to need intensive and tailored support throughout their early schooling in order to stop these gaps from widening or, more hopefully, to help them to close. Providing this support before the critical window for positive early learning closes is likely to offer the best chance of success.

References

[1] Duncan, G. et al. (2007), “School Readiness and Later Achievement”, Psychological Association, Vol. 43/6, pp. 1428-1446, https://doi.org/10.1037/[0012-1649.43.6.1428].supp.

[2] Kautz, T. et al. (2014), Fostering and Measuring Skills: Improving Cognitive and Non-Cognitive Skills to Promote Lifetime Success, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w20749.

Note

← 1. There were too few Pacific Islander/Native Hawaiian and too few American Indian/Alaska Native children in the sample to accurately estimate the skills of these children in IELS.