Early Vocabulary Acquisition: How Word Frequency, Sociodemographic Context, and Intervention Effects Shape First‑Grade Growth

• In Event: The Reduction of Differences in Early Vocabulary

Abstract

Vocabulary is a critical predictor of academic success (Dale et al., 2023), for instance, in reading (Suggate et al., 2018). Children learn words most efficiently when they encounter unfamiliar words frequently and across varied contexts (Hadley et al., 2019). Word frequency, therefore, is a key determinant of word familiarity and thus learning difficulty (Brysbaert et al., 2019). Second language learners (L2-learners) receive less exposure to the language of instruction (Siow et al., 2023), contributing to early, and often widening, lexical gaps between first‑language (L1) and L2 learners (e.g., Chan et al., 2023; Heppt & Stanat, 2020). Systematic, early interventions are needed to narrow these gaps.
The present study thus examines:
RQ1: Does word frequency predict first‑graders’ vocabulary growth?
RQ2: Does frequency operate differently for L1 and L2 learners?
RQ3: Does a digital vocabulary intervention moderate the influence of word frequency and language background on growth?
Method
A sample of 374 German first-graders (58.3% L2-learners; 52.1% female; age: M=78.13 months, SD=4.63) were randomly assigned to one of three intervention groups: Group A (n=157) received vocabulary instruction and practice over 15 weekly units in their regular German lessons, Group B (n=79) received additional treatment during the in-school afternoon program and at home, and Group C (n=138) followed their business-as-usual German lessons. Because preliminary checks showed no difference between Groups A and B, they were pooled (Intervention=236). Word frequency was indexed with ChildLex corpus counts and transformed to Zipf values (Schroeder et al., 2015; Ludewig, 2022). Students were tested at two time points with approximately 18 weeks in between. Vocabulary growth across pre‑ and post‑tests was modeled with explanatory item‑response models that included fixed effects for Intervention, Time, L1/L2 status, and Word Frequency as well as a random effect for Word Item.
Results
For RQ1, analyses revealed that higher‑frequency words were learned more easily (b=0.27, SE=0.11, p=.01), and this advantage did not change over time (Frequency × Time: b= –0.06, SE=0.07, p=.39).
For RQ2, analyses revealed that L2 learners scored well below their L1 peers overall (b=-1.51, SE=0.15, p<.001). The word frequency benefit was also smaller for L2 than for L1 learners (Frequency × Language: b= –0.13, SE= 0.05, p = 0.01), and this gap remained stable across test waves (Frequency × Language × Time: b=0.06, SE=0.07, p=.40).
For RQ3, the digital intervention did not raise overall vocabulary accuracy relative to business‑as‑usual instruction (b= –0.16, SE=0.14, p=.25) and did not alter the frequency advantage for L2 learners (Frequency × Language × Group: b= 0.04, SE=0.06, p=.51). A four‑way interaction that added Time was likewise nonsignificant (b= –0.10, SE=0.09, p=.27), indicating that these null effects persisted throughout the study period.
Discussion
Word frequency remains a robust predictor of early vocabulary acquisition, but L2 learners gain less from high‑frequency exposure than their L1 peers. A digital intervention did not alter this pattern of results, indicating that richer or more sustained instruction may be necessary to reduce early lexical disparities. Selecting target words that balance pedagogical importance with manageable frequency and embedding them in diverse, meaningful contexts may offer a more effective path to narrowing the vocabulary gap.

Authors

• Leonie Dargiewicz, Technische Universität Dortmund

• Svenja Wehrhöfer, TU Dortmund University

• Annika Ohle-Peters, TU Dortmund University

• Ulrich Ludewig, TU Dort­mund University

• Nele McElvany, TU Dortmund University

• Fani Lauermann, University of Bonn