Measuring Elementary School Teachers’ Knowledge of Teaching Vocabulary: A Conceptual Replication and Expansion

• In Event: Investigating Teacher Knowledge and Adaptability: Enhancing Literacy Outcomes in K-3 Students

Abstract

Objectives
This study contributes to the science of reading (SoR) by evaluating and expanding upon an existing measure, the Teacher Knowledge of Vocabulary Survey (TKVS; Duguay et al., 2015), as well as exploring the association between teacher knowledge and student outcomes.

Theoretical Framing
Within one decade, 32 states and the District of Columbia have passed laws requiring that reading instruction be tied to the SoR (Schwartz, 2023). While teacher knowledge can predict student literacy outcomes to varying extents, the term teacher knowledge is broad, encompassing many aspects of the SoR. One such aspect, vocabulary knowledge, is essential for readers to comprehend texts successfully, yet there have been only a handful of measures examining teacher knowledge of vocabulary (Duguay et al., 2015; Phillips et al., 2019) and at the more granular level, morphology (Davis et al., 2021; McMahan et al., 2019). Only one of the four studies mentioned examined associations between teachers’ knowledge and student outcomes. Furthermore, a review of 20 studies focusing on the SoR in teacher preparation and training programs highlighted the need “to gain a more comprehensive picture of the influence of teacher knowledge on student literacy outcomes, especially reading comprehension” (Hudson et al., 2021, p. S311).

Methods
The present study examines the psychometric properties of 26 items from the TKVS and 11 domain-specific vocabulary and morphology items using survey data from 410 third-grade teachers. We then correlate teachers’ knowledge of teaching vocabulary with student domain-specific and domain-general assessments (n = 7,060).

Data Sources
All students had previously participated in the Model of Reading Engagement (MORE) intervention. Additionally, teachers had participated in the state-mandated LETRS training prior to completing the survey. Items from the TKVS were drawn from three sections: Development of Vocabulary Knowledge, Teaching Individual Words, and Fostering Word Consciousness. New items included four questions targeting teachers’ knowledge of domain-specific vocabulary and seven items measuring teachers’ knowledge of domain-specific morphology. Student assessment data consisted of domain-specific comprehension measures from the MORE intervention as well as domain-general standardized measures.

Results
2PL IRT models revealed that items from the TKVS and domain-specific measures had difficulty parameters between -2 and 2.8, although the majority of items had negative difficulty parameters indicating that the items were relatively easy for teachers. Information parameters revealed that on average, most items provided substantial information. Multilevel structural equation modeling indicated that teachers’ knowledge of teaching vocabulary was significantly associated with domain-general (ES=.15) measures but not significantly correlated with domain-specific measures (ES=.05). From a construct validation perspective, the measure correlates with student outcomes. However, including student- and teacher-level covariates eliminates the significant correlation between teachers’ knowledge of teaching vocabulary and domain-general student outcomes.

Scholarly Significance of the Study
Thus, while teachers’ knowledge of teaching vocabulary is important, helping teachers turn that knowledge into engaging classroom instruction is equally important. Instead of targeting teacher knowledge as the silver bullet, the field should consider building teachers’ knowledge of engaging and scientifically-based practices to support literacy instruction. New measures are needed to evaluate these future initiatives.

Authors

• Douglas M. Mosher, Harvard University

• James S. Kim, Harvard University