VoxEU Column Education

Reading to children: a head-start in life

Young children’s cognitive and non-cognitive development significantly affects outcomes for them later on in life. This column asks what effect reading to young children has. Evidence suggests that children should be regularly read to, especially by their parents. Although reading has little effect on non-cognitive skills, the benefits to cognitive development are huge.

The main message from these analyses is that in all approaches and specifications the ‘effect’ of reading to children remains. This ‘effect’ remains over time and spills over into other cognitive skills, particularly those skills more closely related to reading. However, it does not spill over to non-cognitive skills. This consistent effect of reading to children on cognitive skills indicates that it is not just that well-off, well-educated parents read to their children more, generating an association and making it appear as if reading to your child leads to better outcomes for the child. There indeed appears to be a causal effect.

What are the implications of these findings? The main finding is that it is important that young children are being read to. This is an early-life intervention that seems to be beneficial for their early learning outcomes. The study shows that there is an important role for parents in the development of their children. Parental reading to children increases reading and other cognitive skills at least up to the age of 10-11. An interesting further question, which is relevant to policymakers but which cannot be answered with the current data, is whether reading to children at a childcare centre or at school has similar effects.

The cognitive and non-cognitive development of young children is important from an economic perspective because of their effects on economic productivity later on in life (Heckman and Masterov 2007). Cognitive skills are an important determinant in explaining socio-economic success in terms of schooling, wages and quality of jobs. Such skills are influenced by preschool training, education at school but also by parental efforts. Cunha and Heckman (2008), for example, use data from the 1979 US National Longitudinal Survey of Youth to establish the importance of parental investments in raising skills of their children and thus success later on in life. As measures of parental investments they consider the number of books available to the child, whether the child has a musical instrument, whether the family receives a daily newspaper, whether the child receives special lessons and whether the child goes to museums and the theater. They find that the most effective period for cognitive skills investments by parents is early on in the life of their children. Cunha et al. (2006) conclude from an overview of a large number of empirical studies that cognitive ability affects both the likelihood of acquiring advanced training and higher education, and the economic returns to those activities.

Parents reading to their children may stimulate these children to read books themselves and further develop their cognitive skills. Several papers in the education literature have found a positive association of parents reading to their children and the child’s subsequent reading skills, language skills and cognitive development (e.g. see Mol and Bus 2011). However, there is only scant evidence on whether this can be interpreted as a causal effect.

Background

In recent research (Kalb and Van Ours 2013) we examine the effect of parental reading to children early in life on the child’s reading skills using the Longitudinal Study of Australian Children (LSAC). In addition to reading skills at age four to five, reading skills at later ages (up to age 10 to 11) are also examined. At most ages, more than one reading skill measure is observed, which allows for checking the consistency of results when using different measures.

Figure 1. Early reading skill by reading to frequency

The raw data used in our study show patterns indicating a clear association between reading to children more frequently and higher early reading scores (see Figure 1). The graph shows that children who are read to more often are more likely to get a higher score (the lighter coloured parts of the bars). It also shows that girls do slightly better than boys independent of the frequency that they are being read to. This pattern is evident across all measures, with girls doing better than boys in all language-related skills.

For another example of this association, see the skills of girls at age eight to nine, as measured by National Assessment Program – Literacy and Numeracy (NAPLAN) reading tests. Figure 2 shows that girls who are read to more frequently are more likely to score high on the NAPLAN test (i.e., the curve in the graph shifts to the right).

Figure 2. Distribution of NAPLAN reading scores for girls

Methods

In our analysis we use four waves of data of the so-called Child Cohort consisting of just over 4,000 children who are aged four to five years old in the first wave; aged six to seven in wave two; eight to nine in wave three; and 10 to 11 in wave four. Reading skills are measured in a variety of ways: by the parent or teacher at age four to five, by the teacher at age six to seven up to age 10-11, or through NAPLAN. Language skills are measured through the Peabody Picture Vocabulary Test (PPVT) score, and cognitive skills more generally through the learning domain of the Longitudinal Study of Australian Children Outcome Index, as well as non-cognitive skills measured through the socio-emotional domain of the Longitudinal Study of Australian Children Outcome Index. The main variable of interest in this study is the time parents spent reading to the child, which is measured by the frequency with which the child was read to at home, distinguishing between three categories: zero to two times a week, three to five times a week and six to seven times a week. To investigate whether there is a causal effect from reading of parents to reading skills of children we use an instrumental variable approach. The instruments we use are an indicator variable for the oldest child and the number of siblings. These variables are assumed to affect the amount of time parents have available to them to read to their children, but not their children's reading skills directly.

Findings

We have three main findings:

  • Frequency of reading to the child.

For boys, we found that the child is read to less when: the child is older, there are more TVs in the home, more TV is watched on weekdays, and there are more siblings. The child is read to more when: there are more books in the home, the education of either parent is higher, the primary carer is older (up to 50), and the child is the oldest child in household. The results for girls are largely similar, but the effects of age of the parents and the number of siblings are not significant. On the other hand, income now appears somewhat relevant, and watching TV on the weekends is equally important as on weekdays.

  • Factors influencing reading skills.

The reading skills of boys are better when the child is older (within the four to five age range) and a non-English language is spoken at home. Their skills are worse when there are more children books (low significance only) and the primary carer is older (up to age 40). Similar results are observed for girls, except that the number of TVs now has a positive effect (low significance), which is counteracted by a negative effect from the amount of TV watched on weekdays, and the education of the primary carer now has a small negative effect on the reading skill (again low significance). Although the education of the parent has no effect (or a small effect opposite to what is expected) on reading skill at age four to five, positive effects of parental education are estimated for later reading and other skills.

  • Effect of reading to the child.

We find that not only is there a strong association between reading to children and the child’s early reading outcomes, we also find that this effect is causal. Moreover, our results indicate that, if anything, the causal effects of reading to children are larger than the observed associations in the raw data, which show a difference of 0.26 standard deviations of the skill measure between a child that is read to six to seven times per week and a child that is read to 0-2 times per week.1 Figure 3 shows an overview of our results. The top bars clearly show a larger effect than the raw association at more than 0.5 standard deviations. To place the size of these effects in context, they can be compared to the effect of age. For boys, reading six to seven days per week (compared to two or less) has a slightly larger effect than being six months older. The effects for girls are larger relative to age.

A series of robustness tests (many presented in Figure 3) confirm the significance and direction of the effects. Checks included using different Longitudinal Study of Australian Children data from the so-called Birth cohort; examining the effect of reading to children at an earlier age; using outcomes at a later age and for different skills; and using alternative methodologies, such as the propensity score matching approach. In addition, in alternative specifications: other parental activities with the child were added to allow for the effect of time spent with the child through other activities than reading from a book; birth characteristics were added to check for the effect of being the oldest child working through biological differences rather than the extra time spent with the parents; the number of siblings at age four to five is replaced with the number of siblings as measured at a later point in time to check whether the effect might work through identifying the socio-economic status of the family rather than time constraints. All these alternative explanations were rejected.

The results in Figure 3 point to the effects of reading to a child being stronger for outcomes that reflect the child's own reading levels, than for outcomes on other cognitive measures, such as numeracy. The magnitudes of the effects are large, with the exception of effects on non-cognitive skills, which disappear in most specifications that control for the bias in the ‘reading to’ variable.

Figure 3. Effect of reading to children at age four to five (six to seven times per week versus less than twice per week) on a range of skill measures expressed in standard deviations of the measure

Policy implications

The main message from these analyses is that in all approaches and specifications the ‘effect’ of reading to children remains. This ‘effect’ remains over time and spills over into other cognitive skills, particularly those skills more closely related to reading. However, it does not spill over to non-cognitive skills. This consistent effect of reading to children on cognitive skills indicates that it is not just that well-off, well-educated parents read to their children more, generating an association and making it appear as if reading to your child leads to better outcomes for the child. There indeed appears to be a causal effect.

What are the implications of these findings? The main finding is that it is important that young children are being read to. This is an early-life intervention that seems to be beneficial for their early learning outcomes. The study shows that there is an important role for parents in the development of their children. Parental reading to children increases reading and other cognitive skills at least up to the age of 10-11. An interesting further question, which is relevant to policymakers but which cannot be answered with the current data, is whether reading to children at a childcare centre or at school has similar effects.

References

Cunha, F, JJ Heckman, LJ Lochner and DV Masterov (2006), “Interpreting the evidence on life cycle skill formation”, in: Hanushek, EA and F Welch (eds.) Handbook of the Economics of Education, Amsterdam, Elsevier, 697–812.

Cunha, F and JJ Heckman (2008), “Formulating, identifying and estimating the technology of cognitive and noncognitive skill formation”, Journal of Human Resources 43, 738–782.

Heckman, JJ and DV Masterov (2007), “The productivity argument for investing in young children”, Review of Agricultural Economics 29(3), 446–493.

Kalb, G and JC van Ours (2013), “Reading to children gives them a head-start in life”, CEPR Discussion Paper 9485, May.

Mol, SE and AG Bus (2011), “To Read or Not to Read: A Meta-Analysis of Print Exposure From Infancy to Early Adulthood”, Psychological Bulletin, 137, 267–296.


1 A standard deviation is the average deviation from the mean value, indicating how much variation there is in individual outcomes. Using this approach enables comparison of effects across skill measures with different measurement units.

8,294 Reads