Vocabulary interventions for second language (L2) learners up to six years of age

Background: Second language (L2) learners are a heterogeneous group. Their L2 skills are highly varied due to internal factors (e.g. cognitive development) and external factors (e.g. cultural and linguistic contexts). As a group, their L2 vocabulary skills appear to be lower than their monolingual peers. This pattern tends to persist over time and may have negative consequences for social interaction and inclusion, learning, and academic achievement.

Objectives: To examine the immediate and long-term effects of second language (L2) vocabulary interventions targeting L2 learners up to six years of age on vocabulary and social-emotional well-being. To examine the associations between L2 vocabulary interventions and the general characteristics of L2 learners (e.g. age, L2 exposure, and L1 skills).

Search methods: We used standard, extensive Cochrane search methods. The latest search date was December 2022.

Selection criteria: We included randomised controlled trials (RCTs) comparing the effects of vocabulary interventions for L2 learners up to six years of age with standard care.

Data collection and analysis: We used standard methodological procedures expected by Cochrane. Our primary outcomes were 1. receptive and 2. expressive L2 vocabulary (both proximal and distal), and 3. mean length of utterance (MLU; which is a measure of potential adverse effects). Our secondary outcomes were 4. L2 narrative skills, 5. L1 receptive vocabulary (both proximal and distal), 6. L1 expressive vocabulary (both proximal and distal), 7. L1 listening comprehension, 8. L2 grammatical knowledge, 9. L2 reading comprehension (long-term), and 10. socio-emotional well-being (measured with Strengths and Difficulties Questionnaire).

Main results: We found 12 studies involving 1943 participants. Two studies were conducted in Norway, seven in the USA, and single studies conducted in Canada, China, and the Netherlands. Ten studies were conducted in preschool settings, with a preschool teacher being the most common delivery agent for the intervention. The interventions were mainly organised as small-group sessions, with three or four children per group. The mean dosage per week was 80 minutes and ranged from 24 to 120 minutes. The studies commonly applied shared book reading (reading aloud with the children), with target words embedded in the books. Standard care differed based on the setting and local conditions in each country or (pre)school. In some studies, the comparison groups received vocabulary instruction in preschool groups. Compared to standard care, the effect of L2 vocabulary interventions varied across outcome measures. For vocabulary measures including words that were taught in the intervention (proximal outcome measures), the intervention effects were large for both receptive L2 vocabulary (i.e. understanding of words; standardised mean difference (SMD) 0.97, 95% confidence interval (CI) 0.64 to 1.30; 4 studies, 1973 participants; very low-certainty evidence) and expressive L2 vocabulary (i.e. expressing or producing words; SMD 0.86, 95% CI 0.56 to 1.17; 6 studies, 1121 participants; very low-certainty evidence). However, due to some concerns in the overall risk of bias assessment, substantial heterogeneity, and wide CIs, we have limited confidence in these results. For language measures that did not include taught vocabulary (distal outcome measures), the intervention effects were small for receptive vocabulary (SMD 0.29, 95% CI 0.02 to 0.55; 6 studies, 1074 participants; low-certainty evidence) and probably made little to no difference to expressive vocabulary (SMD 0.10, 95% CI -0.02 to 0.23; 7 studies, 960 participants; moderate-certainty evidence). There was little to no intervention effect on L2 listening comprehension (SMD 0.19, 95% CI -0.31 to 0.68; 2 studies, 294 participants; very low-certainty evidence), but the evidence was uncertain, and the interventions probably increased L2 narrative skills slightly (SMD 0.37, 95% CI 0.14 to 0.59; 2 studies, 487 participants; moderate-certainty evidence). Only one study reported data on MLU, and we were unable to examine the effect of intervention on this outcome. The level of certainty of the evidence was downgraded mainly due to inconsistency and imprecision. We were unable to draw conclusions about socio-emotional well-being, or conduct the planned subgroup analyses to examine the second objective, due to lack of data.

Authors' conclusions: Findings from this review suggest that, compared to standard care, vocabulary interventions may benefit children's L2 vocabulary learning but have little to no effect on their listening comprehension, though the evidence is uncertain. Vocabulary interventions probably improve the children's storytelling skills slightly. Due to the limited number of studies that met our inclusion criteria and the very low- to moderate-certainty evidence as a result of inconsistency and imprecision, implications for practice should be considered with caution. This review highlights the need for more high-quality trials (e.g. RCTs) of vocabulary interventions for L2 learners, particularly studies of learners outside the USA.