A Systematic Review of Word Selection in Early Childhood Vocabulary Instruction

• In Event: Disrupting the Status Quo in Vocabulary: Radically Rethinking "What," "Who," and "How" for Vocabulary Instruction

Abstract

Over the past several decades, a rich research base has emerged on how to support vocabulary in early childhood classrooms (Marulis & Neuman, 2010; Mol, Bus, & de Jong, 2009). One understudied aspect of early childhood vocabulary interventions concerns the selection of words for instruction. Choosing the right words to teach is vital in supporting children’s ability to access and understand text. Teaching words that are too esoteric (e.g., gunwale), or are already known (e.g., puppy) may have little impact on general vocabulary knowledge or on reading comprehension. The type of words taught also significantly influences how much children learn about new words, and the magnitude of effect sizes (Authors, 2016). Therefore, a closer examination of which words that are taught in vocabulary interventions is important in understanding what the field knows, and has yet to learn, about vocabulary instruction.
The purpose of the present study is to systematically review the words selected for instruction (“target words”) in pre-k and kindergarten vocabulary intervention studies from the last twenty years to better understand and contextualize the learning reported by these studies. Specifically, we examine the methods used to select words, target word characteristics such as concreteness and part of speech, target words' difficulty and usefulness, the context from which words were drawn, and the relationship between child and word characteristics.
We used systematic review methods to search for studies relevant to our search criteria, searching major databases and hand-searching key journals and authors. All abstracts (n = 3,227) were double-screened for inclusion and exclusion criteria by the first and second author. We included studies with preschool and kindergarten child participants in a classroom setting in which children were taught new vocabulary words. We excluded studies in which children had diagnosed developmental disabilities and in which the study took place in a non-English-speaking context (for comparability across target words). The first and second author double-coded all studies that qualified for inclusion (n = 76) for study, participant, target word, and intervention characteristics.
Analyses revealed great variability in target word selection and inconsistent adherence to research-based principles of word selection. Researchers prioritized unknown words and words described as Tier 2 (Beck, McKeown, & Kucan, 2013) when selecting words. The most frequently selected word type was concrete nouns; more than half of all words selected were concrete in nature (see Figure 1). Only a small percentage of target words appeared on published lists of words recommended for instruction (see Table 1). Word selection varied by context, with different types of words selected for science content vs. narrative/fiction. Finally, there were few relationships between child characteristics and word characteristics. For example, there were no differences in the frequency or difficulty of words chosen for children who were dual language learners (DLLs) vs. English-only learners. Implications for vocabulary instruction and research will be discussed, such as the importance of choosing words that meet standards for ecological and social validity.

Authors

• Elizabeth Burke Hadley, University of South Florida - Tampa

• Karyn Mendez, University of South Florida