Fostering language skills for children in less-educated households: Evaluating a parental reading initiative
Skill-development processes in the preschool years are of pivotal importance for educational success (Heckman 2019). Core language skills such as vocabulary, phonemic awareness, and knowledge of print are strong predictors of academic performance. Unfortunately, research shows that children from less-educated and immigrant families are highly overrepresented among low achievers in the preschool years, with long-term consequences for their school success. Public investments to foster language skills in early childhood and reduce related social inequalities can thus be justified on both efficiency and equity grounds.
Several scholars have advocated a social investment strategy promoting access to high-quality educational services in early childhood. While these school-based interventions can be useful, they tend to leave the family context virtually unchanged. Hence, parenting interventions have been increasingly advocated as a complementary approach to boost early skill accumulation and reduce educational inequality. Informal learning processes that take place in the home environment, such as Shared Book Reading (SBR), are strong predictors of early language skills and school success. Since these informal activities are socially stratified, interventions broadening access to effective parenting practices also have the potential to reduce related social inequalities. Moreover, parenting programmes provide a light-touch, cost-effective complement to school-based interventions, which can be expensive.
Indeed, fostering SBR in early childhood is regarded as a promising approach to boost parental involvement in disadvantaged contexts. This activity demands limited time investments from parents and is accessible to families with few socioeconomic resources, since free or low-cost storybooks for children are easy to find and use a simple language register. Unsurprisingly, educational experts recommend SBR as a beneficial parenting practice. Large-scale home-visiting interventions, such as Bookstart, have been developed in several countries to encourage this practice by providing free books to parents as well as information on the beneficial effects of SBR for children. Paediatric interventions have become increasingly popular as well. The best-known example is the ‘Reach Out and Read’ model, whereby paediatricians provide free books and information on the benefits of SBR to parents.
The challenges of boosting effective parenting practices
Still, boosting effective parenting practices remains a major challenge. Systematic reviews of parental involvement interventions report mixed results, suggesting that economic, cultural, and social barriers may hinder parental receptivity to these interventions (Gorard and See 2013). Barriers include parental distress resulting from economic insecurity, family instability, lack of familiarity with books and low reading skills, limited investment focus from parents, and language barriers for immigrant parents.
The consensus on the efficacy of SBR interventions that emerged in the 1980s and 1990s was based on correlational research designs. However, an increasing number of randomised experiments has recently reported more elusive evidence. A meta-analysis of 30 experiments on the impact of SBR interventions on children’s vocabulary reports that most programmes fail to make a significant impact (Barone et al. 2019). Only one SBR methodology – namely, dialogic reading – is associated with systematically positive effects. In particular, the mean size of the effect of SBR interventions not involving dialogic reading is very close to null (0.06). Moreover, this meta-analysis reports that treatment effects tend to fade out a few months after the intervention. Treated children may rapidly forget the new words they learned, while children in the control group learn them even without participating in SBR interventions. While the evidence for dialogic reading is encouraging, the approach is less effective among poor families (Mol et al. 2008, Barone et al. 2019a). Hence, its efficacy comes at the price of increasing socioeconomic gaps in early language skills. Developing effective and equalising SBR interventions remains a major challenge.
These experiments are mostly based on small samples (106 cases, on average) and could thus fail to detect treatment impacts. Moreover, the recourse to ad hoc sampling methods weakens the generalisability of their results. We carried out a randomised experiment based on a sample of 1,700 children, aged four, living in disadvantaged neighbourhoods in Paris and attending pre-primary schools (Barone et al. 2019b). The intervention enhanced SBR frequency in the treatment group, with positive effects on children’s vocabulary as measured via standardised test scores. These effects were fully persistent at a follow-up carried out six months after the intervention. Crucially, the intervention was more effective among less-educated families, and it was as effective among immigrant families as it was among native families. What made a difference relative to previous interventions? We suggest that a strong focus on the accessibility of treatment contents for disadvantaged parents is a key ingredient. Below, we explain how we tried to identify and address the hurdles that can hinder receptivity to SBR interventions for less-educated immigrant parents.
Fostering the accessibility of SBR interventions
SBR interventions are based on the premise that parents may not be aware of the activity’s potential benefits. These information gaps are particularly strong among less-educated parents, which motivated our decision to develop an information initiative. However, effectively reaching parents to deliver this information is challenging, particularly in low-income schools with limited parental involvement. Our pilot study indicated that a face-to-face intervention (e.g. via a school meeting with parents) would result in low attendance and a strong bias towards more educated parents. We thus decided to deliver the main information contents with one flyer per week that used simple language and could be read in two minutes. However, assuming that some parents would not read the flyers, we hired four interviewers to reach each family with a short phone call summarising the main contents of the flyers and encouraging parents to read them.
Language barriers for immigrant parents are another important barrier. We adopted two expedients. First, the phone calls could be delivered in four languages (French, Arabic, English, or Spanish). Second, we selected books with limited text and several images that parents could use to tell the story even without reading the full text.
Even parents who receive accurate information on SBR and appreciate its value may fail to incorporate it in their parenting routines. For instance, in less-educated families, books for children may not be available at home and parents might not borrow them from public libraries. We thus organised a school-based book loan: teachers received a selection of 18 titles and children could choose one book per week. The pilot study had indicated the importance of letting children choose their books. When children are interested in a book, they ask parents to read it, which is the best way to persuade parents to take the time to do so. Children thus came home with a project bag containing the weekly flyer and a book. Moreover, in order to keep parents involved and mobilised, we sent a weekly text message during the third and the fourth month of the intervention. Hence, for four months, families received storybooks together with regular information stimuli via flyers, phone calls, and text messages. A potential limitation of some existing SBR programmes, such as Bookstart, is their reliance on one-shot interventions.
The contents of these messages involved two main clusters. A first set of messages involved the importance of early language development for school success. The information materials explained to parents that reading storybooks on a regular basis can enrich children’s vocabulary, which is in turn essential to reading and writing in primary education. A second set of messages focused on the quality of SBR activities. A more interactive reading style is more beneficial to emergent literacy, and so we provided parents with practical tips to encourage their children to talk about the story. Moreover, we explained how to effectively ‘play the story’ using voice and gesture to mimic the behaviour of characters and express their emotions.
The books selected for SBR interventions are a relevant element of treatment design. In particular, the balance between accessibility and vocabulary enrichment is a delicate yet essential component. If a book contains several unfamiliar words for children, they will have a hard time following the story; but if the language is too simple, they will not learn new words. Moreover, research based on video recordings of SBR reveals that parents tend to exclusively focus on the text, while children are attracted to the book’s images and can stay quite disconnected from its verbal contents. Books that do not contain long portions of text but are rich in images – while displaying clear connections between the text and images – are more likely to foster engaged interactions. The materials used in this SBR intervention are freely available and are currently used by pre-primary schools.
Barone, C, E Chambuleyron, R Vonnak and G Assirelli (2019a), “Shared Book Reading Interventions and Children’s Skills: a Meta-Analysis of Randomised Controlled Trials”, LIEPP working paper.
Barone, C, D Fougère and C Pin (2019), “Social origins, shared book reading and language skills in early childhood: evidence from an information experiment”, CEPR Discussion Paper 14006.
Gorard, S and B See (2013), “Do parental involvement interventions increase attainment?“.
Heckman, J (2019), “There’s more to gain by taking a comprehensive approach to early childhood”.
Mol, S, A Bus, M De Jong and J Smeets (2008), “Added value of dialogic parent-child book readings: a meta-analysis”, Early Education and Development 19(1): 7-26.