Language Development and Enjoyment of Reading: Impacts of Early Parent-Child Activities in Two Growing up in Scotland Cohorts

Chapter 1: Introduction

This report uses data from the two birth cohorts in the Growing Up in Scotland study which are representative of children aged 10 months and living in Scotland in 2004/05 and 2010/11. Data from both cohorts is used to compare language development at age 3 and whether there is any evidence that this is linked to changes in early parent-child activities across the two cohorts. A secondary question explored in the report is whether any changes in home learning activities across the cohorts may be linked to the introduction of the Bookbug programme and the PlayTalkRead campaign - the timing of which meant that only Birth Cohort 2 would have been exposed to these. Finally, data from children themselves in the older cohort is used to examine the relationship between early parent-child reading and enjoyment of reading at age 8.

1.1 Growing Up in Scotland ( GUS)

Growing Up in Scotland ^[1] is a longitudinal research study tracking the lives of thousands of children and their families in Scotland from the early years, through childhood and beyond. The main aim of the study is to provide new information to support policy-making in Scotland but it is also intended to provide a resource for practitioners, academics, the voluntary sector and parents. To date, the study has collected information about three 'cohorts' of children: a child cohort and two birth cohorts - altogether, information has been collected about 14,000 children. The child cohort included 3000 children born between June 2002 and May 2003. In total, four 'sweeps' of data were collected from these families: first when the children were aged just under 3, and then annually until the children were just under 6. The first birth cohort ( BC1) comprised around 5000 children born between June 2004 and May 2005. For this cohort, data was collected annually from when the children were aged 10 months until they were just under 6 years old, and then at age 8 and during the time children were in Primary 6. ^[2] The second birth cohort ( BC2) comprised approximately 6000 children who were born between March 2010 and February 2011. For this cohort, data has been collected when the children were aged 10 months, just under 3 years, just under 4 years, and just under 5 years. ^[3]

GUS has collected data on a wide variety of issues related to early child experiences, including the child's experience of reading (or being read to) and playing, but it also includes data from specific assessments of the child's oral language abilities at 3 years of age. When children in BC1 were around 8 years old (at sweep 7), they were asked to complete a short self-complete questionnaire as part of the data collection exercise. This included questions on whether they enjoyed reading. It is thus possible, using GUS data, to explore the relationship between the child's early exposure to reading (either reading or looking at books themselves or being read to) and their language ability at age 3, as well as to explore any relationship between their early reading and later enjoyment of reading.

1.2 PlayTalkRead and Bookbug

Between the birth of children in the two cohorts the Scottish Government supported the introduction of two national interventions aimed at improving parents' access to information and resources on parent-child activities in the early years: the PlayTalkRead campaign and the Scottish Book Trust's Bookbug programme. PlayTalkRead and Bookbug both contribute to the Scottish Government's early years agenda. The initiatives are a key part of the commitment to early intervention initially set out in the Early Years Framework, Equally Well, and Achieving Our Potential (Scottish Government 2008a, 2008b, 2008c), and more recently in policies such as the National Parenting Strategy (Scottish Government, 2012) and the Play Strategy (Scottish Government, 2013).

PlayTalkRead is a Scottish Government campaign launched in 2009 which aims to support and encourage parents to stimulate their children from an early age through playing, talking and reading with them on a regular basis. The campaign seeks to achieve this through providing parents and carers of children up to 3 years' of age with free or low cost ideas of how to positively engage with their children in fun and playful ways. It has a particular focus on reaching parents in the most socio-economically disadvtanged groups (C2DE2). Building on research evidence on child development, the campaign seeks to highlight the importance of play and playful interaction for boosting children's development within a number of areas including communication, literacy, motor and problem solving skills development (Working on Wheels, 2015). The campaign has been delivered through a combination of TV, outdoor, online advertising, social media, PR, partnerships, a website, and two PlayTalkRead buses (PlayTalkRead, 2015).

Two key elements of the campaign are the PlayTalkRead website and two PlayTalkRead buses that seek to bring the campaign to local communities across Scotland. Both elements are outsourced. The PlayTalkRead website contains hints, tips and ideas for stimulating children in cost-effective ways. Re-vamped in early 2011 and again in 2014, the website contains digital books and interactive videos, and parents can register for an online community where they can share experiences with other parents (Omerod, 2011).

The buses are an important element of the PlayTalkRead campaign in that they go into communities and provide play areas for young children and their parents or carers, with play workers available to support and encourage activities. From 2012, the purpose of the PlayTalkRead buses expanded to focus not only on providing play facilities but to also actively support parents and children to play through face-to-face engagement. Furthermore, from April 2012 onwards, there was an increased focus on ensuring that buses visited areas with high levels of deprivation ^[4] .

Bookbug is the Scottish Government's early years book gifting programme; a universal book gifting scheme run by the Scottish Book Trust. The programme seeks to promote the importance of books and the benefits of early book sharing, and an important aim of the programme is to encourage parents to share books with their child or children from an early age. In addition to laying the foundations of early literacy, Bookbug aims to improve attachment between young children and their parents or carers, as well as to increase children's emotional intelligence, communication and listening skills. When launched in 2010, Bookbug was primarily concerned with gifting book packs to babies, toddlers and ante-preschool children (age 3 years). Packs were distributed to all families in Scotland, typically through their health visitor or early years setting. In addition to this, free song and rhyme sessions were held, in local libraries and public spaces. Since 2010 the programme has expanded and a further book pack is now gifted to children when they start primary school. Furthermore, Bookbug now also involves 'assertive' outreach work targeted at children in disadvantaged circumstances (Scottish Book Trust, 2012), the main element of which is 'Bookbug for the Home' which involves trained practitioners taking elements of the Bookbug approach into family homes. Beginning in 2012, this was rolled out in 8 local authorities each year and is now available in all 32 local authorities.

Bookbug is a follow-up to Bookstart, a book gifting programme which started as a local project in Birmingham in 1992 and was subsequently rolled out across the UK. While Bookstart is still active in England and Wales, in Scotland it was replaced by Bookbug in 2010. Bookbug was developed as a specifically Scottish book gifting programme which featured more Scottish authors and illustrators and tied in with Scottish education and parenting policies.

Like Bookbug, Bookstart in Scotland consisted of a mixture of universal book gifting, free song and rhyme sessions, and some outreach work. Thus, prior to the roll out of Bookbug in June 2010, children in Scotland received at least one Bookstart book bag in their first years of life, and Bookstart Rhymetime sessions would have been available across the country.

1.2.1 The significance of the timing of BC1 and BC2

Because the same questions were asked of the parents of both cohorts and the same measure of language was used at 3 years it is possible to explore whether children in BC2 had better language skills than those in BC1 and whether their parents read to them more. Furthermore, because Bookbug and PlayTalkRead were introduced shortly before the children in BC2 were born (but not until BC1 had passed through their pre-school years), GUS gives us an opportunity to explore whether there is any evidence that changes in home learning activities across the cohorts might be linked to the introduction of these initiatives.

At their first interview, all BC2 respondents were asked whether they remembered having received a Bookbug pack, and to what extent they had made use of the different elements within the pack. In addition, parents in BC2 were also asked whether they had used the PlayTalkRead website when their child was 10 months and 3 years old, respectively. It should be noted that neither the questions asked in GUS, nor this analysis, were designed to be an evaluation of Bookbug or PlayTalkRead. The Bookbug scheme included free song and rhyme sessions and has since expanded to include outreach work targeted at children in disadvantaged circumstances. However, the questions asked in GUS allowed us to establish only whether parents recalled receiving and using the initial Bookbug pack issued when their child was a baby. The PlayTalkRead campaign has been delivered through a combination of TV, outdoor, online advertising, social media, PR, partnerships, a website, and a roadshow element, but the questions asked in GUS focused only on whether the parent engaged with one element of the initiative - the website. Nevertheless, GUS does present an opportunity to explore the relationship between home learning practices and these specific elements of the inititiatives.

1.3 Research questions

Within each of the topics being considered - early language development, home learning activities, use of interventions, and enjoyment of reading - this report aims to answer a number of distinct research questions.

1.3.1 Early language development and home learning acitivities

The research questions related to early language development and home learning activities are as follows:

• Was there an improvement in language development across the two cohorts?
• Did any improvement remain significant after taking account of other differences in family background between the cohorts that would have affected language development, especially parental level of education?
• Were there any changes in language development within particular socio-economic groups? For example, did the difference between the most and least advantaged groups change?
• Was there an increase in home learning activities across the cohorts overall?
• Did any increase remain statistically significant after taking account of other differences in family background between the cohorts that would have affected home learning, especially parental level of education?
• Were there any changes in the nature or frequency of home learning activities between cohorts within particular socio-economic groups? For example, did the difference between the most and least advantaged groups change?
• In BC2, was there an independent relationship between home learning activities and language development similar to that previously found in BC1 data?
• Did the strength of this relationship vary for different socio-economic groups?
• Was the relationship different ( i.e. stronger or weaker) in BC2 than it was in BC1?
• Do early learning activities reduce the negative effects of disadvantage on language development

Consideration of language development initially explores the difference in language ability at age 3 between children in each cohort; first at an overall level and then amongst children with different social backgrounds.

The frequency of home learning activities is then explored with a particular focus on examing whether this has changed between cohorts both at an overall level and within sub-groups of parents with different levels of education. Change within sub-groups in this, and other sections, is considered in order to establish whether the relationship between social background and participation in home learning activities (or language development) has changed. In particular, whether the difference between those in the most and least advantaged groups has decreased.

1.3.2 Use of interventions

In relation to use of the Bookbug and PlayTalkRead interventions, the report seeks to answer the following questions:

• Did parents in BC2 receive and make use of the PlayTalkRead website and of the first Bookbug pack and did this use vary across different socio-economic groups?
• In BC2, was parents' receipt and use of the first Bookbug pack and having accessed the PlayTalkRead website associated with increased participation in home learning activities?
• In BC2, was parents' receipt and use Bookbug and having accessed the of PlayTalkRead website associated with children's language ability?

This analysis begins by considering knowledge and use of the first Bookbug pack and the PlayTalkRead website amongst parents in the younger cohort ( BC2). As well as illustrating the overall proportion using these resources, differences in use by social background characteristics are also explored. Social background characteristics considered include parental level of education, annual household equivalised income and area deprivation.

The analysis then examines - for BC2 parents - whether, after controlling for background characteristics, parents who report accessing the PlayTalkRead website and/or using the first Bookbug pack showed differences in the activities they undertook with the child.

For BC2 parents, analysis is conducted to determine whether receipt and use of the Bookbug pack and accessing the PlayTalkRead website was independently associated with improved language ability after controlling for other potentially influencing factors.

The conclusions discuss the extent to which:

• Home learning activities could be improved/increased amongst parents and amongst which groups such improvement may lead to the greatest positive results
• The interventions, as measured in GUS, appear to play a role in doing that.

1.3.3 Enjoyment of reading

When considering enjoyment of reading, the questions we sought to answer were:

• How did enjoyment of reading vary amongst children at age 8?
• Was there an independent relationship between early parent-child reading and later enjoyment of reading? In other words, after controlling for other factors which may influence enjoyment of reading, did being read to in the early years increase the likelihood that children enjoyed reading at age 8?
• Did the strength of this relationship vary for different socio-economic groups?

In answering the first question, the report provides a descriptive overview of differences in children's enjoyment of reading. It explores variations according to key social and demographic characteristics such as gender and household income, as well as variations in other aspects of the child's life including enjoyment of school and other activities such as number work, sports and games and learning.

The second and third questions involve examining the relationship between exposure to reading in the early years (this has been measured annually from birth to age 6) and enjoyment of reading at age 8, as reported by the child. This shows whether children who are read to more often in their early years are more likely to enjoy reading at age 8, and whether the relationship between early reading and later enjoyment is similar for children with different background characteristics. In considering these analyses, the conclusion will consider:

• Whether some of a child's later enjoyment of reading is due explicitly to having been read to when younger. Thus, do more children from advantaged circumstances enjoy reading at age 8 because they were more likely to be read to when younger? And are children from more disadvantaged backgrounds who were read to more often when younger more likely to enjoy reading than their peers who were not read to?
• If all children had the same early exposure to reading would children from more advantaged groups still be more likely to enjoy reading? In other words, does social background continue to have a direct effect on enjoyment of reading after controlling for differences in early exposure to reading?
• Whether the effect of early exposure to reading varies for children with different background circumstances? For example, is the relationship between early reading and later enjoyment stronger for children in more advantaged groups than in more disadvantaged groups ( i.e. do they benefit more from it) or is there no difference?

1.4 What we already know from GUS about early language development and its relationship with home learning

Earlier reports from GUS have demonstrated stark variation in language development at ages 3 and 5 amongst children from different backgrounds. For example, Bromley (2009) found that at age 3, children from less advantaged families were outperformed by their more affluent counterparts and noted significant differences in ability according to, amongst other things, gender, maternal age, family composition, early development and birth weight. Bradshaw (2011) found these inequalities largely persisted at age 5. In these and other reports early language development has been shown to be associated with a wide range of factors present in children's lives including childcare and pre-school experience (Bradshaw et al., 2014; Bradshaw and Wasoff, 2009), maternal mental health (Marryat and Martin, 2010) and experience of poverty (Barnes et al., 2010). Of particular significiance, as far as the present report is concerned, several reports have demonstrated links between early home learning activities and language development (Bradshaw et al., 2014; Bradshaw 2011; Melhuish, 2010; Bromley, 2009). Indeed, participation in such activities - which include reading with the child, painting or drawing, and singing nursery rhymes - on a regular basis have been shown to be an important protective factor associated with better language development amongst children from more disadvantaged backgrounds.

1.5 The evidence base underpinning early book reading interventions

A great many studies have looked at the effects of different aspects of book reading. For example, dialogic book reading interventions, where parents are explicitly taught to actively engage their children in reading activities, have been shown to promote oral language and literacy skills and to be an important element in the pathway to school readiness, literacy and attainment more generally (Crain-Thoreson and Dale, 1992). Indeed, active parental reading has been found to mediate almost all of the advantage in early language outcomes associated with higher levels of maternal education (Farrant and Zubrick, 2011). Examples of book reading interventions which have been identified by the Early Intervention Foundation ( http://www.eif.org.uk/) as having a reasonable underpinning evidence base include Parents Early Education Partnership ( PEEP) (Evandelou and Silva, 2003), Reach Out and Read (Sharif et al., 2002), Raising Early Achievement in Literacy ( REAL) (Hannon, Nutbrown and Morgan, 2005), Hear and Say (Heubne and Meltzoff, 2005), Bookstart (Wade and Moore, 2000), Bookstart Corner (Demack et al., 2013) and Bookstart Plus (O'Hare and Connolly, 2010). Of particular interest is the role played by interventions focusing on promoting early dialogic book reading ^[5] with young children:this not only encourages familiarity with the process of looking at books and by extension reading them but also provides a structure to enhance parent-child interaction. The single most commonly cited review of the evidence (Bus et al., 1995) is a meta-analysis of the effects of targeted interventions to promote parental reading on language growth, emergent literacy and reading achievement. The overall effect size is relatively high and has triggered an understanding that the outcomes of interventions to promote dialogic book reading are positive. However, it is not clear what the key ingredients of such programmes are, how applicable they are as universal interventions, and whether there is a threshold in terms of the quality of intervention which needs to be reached before change can be expected.

Although dialogic book reading interventions have been a key feature of early interventions, there has also been a move towards book gifting programmes. This involves distributing children's books to families when the child is born, or soon afterwards, together with general advice about reading to the child but generally without any direct instruction. This practice grew out of a finding that the number of books in the home was indicative of later attainment. Simply having the books was thought to make a difference. Indeed, bookgifting programmes such as Bookstart became a standard component of many Surestart local programmes in the UK. Books were given to families with young children and sessions were provided in local libraries and children's centres. Yet rarely was explicit reference made to the literature in the planning of these services and where it was, the outcomes were often unclear. In a recent literature review by Burnett et al. (2014), commissioned by the Booktrust, the authors explore the impact of bookgifting programmes on literacy attainment more widely. The review is structured according to three dimensions which were identified as being promoted through Bookstart. The three dimensions are: reading for pleasure, book ownership and regular book sharing from an early age. The review's authors stress that while some studies have reported positive impacts on a number of measures, other studies have not. They reference a number of studies which found that parents who had participated in a book gifting programme reported reading more frequently and/or spending more time reading with their child(ren) than parents who had not participated in a book gifting scheme. A note of caution is provided, however, suggesting that the relationship between book gifting programme participation and book sharing frequency is likely to be stronger amongst families who engaged less in book sharing prior to participating in the bookgifting programme (Burnett et al., 2014:36).The authors conclude '…we would suggest that whilst evidence on the impact of bookgifting is mixed, there is promising evidence that bookgifting is linked to later improvements in reading. We see the strongest evidence of this in evaluations focused on populations who have had little experience of booksharing in the past. Mounting evidence from evaluations suggests that bookgifting programmes can impact on: parental attitudes to sharing books with young children; children's enthusiasm for looking at books; the frequency and extent of booksharing; and book ownership and library membership.'

1.5.1 Evaluating early book reading interventions

Book gifting and these more general exhortations to parents to read more to their children tend to be "universal" interventions in the sense that all families are given books irrespective of any specific identified need, and the messages are readily available through publicity and social marketing campaigns, with many available on the internet etc. In order to evaluate the effectiveness of such programmes it is necessary to examine them in whole populations and to establish whether the effects are sustained over time. This can be difficult to achieve because of the potential scale of such studies but also because only very rarely are families followed up over time to see whether a given programme has had any lasting effect. The BC1 and BC2 cohorts in GUS arguably provide such an opportunity in two respects. First, the timing of the roll-out of the PlayTalkRead and Bookbug initiatives means that many families in BC2 were exposed to both initiatives, while BC1 families were not exposed to either (although BC1 families may have received packs from Bookbug's predecessor, Bookstart). Of course, comparing cohorts in this way does not establish a causal relationship between the availability of the programme materials and any changes in the cohorts, but it does allow us to establish what sort of changes are taking place over time. Second, because parents in BC2 were asked questions both about their use of the first Bookbug pack and one aspect of the PlayTalkRead (the website) and about the activities they undertook with their child, BC2 data itself enables an exploration of whether there was any association between engagement with these aspects of the initiatives and the frequency of home learning activities undertaken.