Empirical Design Parameter Values for Planning Mediation Studies With Teacher and Student English Language Arts Outcomes

• In Event: Methodological Research in Education

Abstract

In the field of education research, the use of mediation analyses to test and unpack program
theories is prevalent and increasingly adopted by applied researchers. However, there is a lack of
empirical literature that provides guidance on how to effectively and efficiently design such
mediation studies. To bridge this gap, our study aims to contribute by providing initial empirical
estimates for design parameter values in mediation studies. In this paper, we focus on literacy
(English language/arts) teacher mediators and their impact on student literacy outcomes in a
three-level structure (students within teachers within schools). Our analysis goes beyond
previous research by investigating a wide range of contexts (e.g., academic, summer programs),
data collection methods (including surveys, observations, and assessments), types of mediators
(knowledge, instruction, affect, beliefs, perceptions, attitudes, climate, program implementation),
measurement instruments (teacher knowledge for teaching literacy, CLASS), and domains (e.g.,
comprehension, word analysis, and classroom management). The findings highlight the
importance of considering design parameter estimates specific to the mediator and outcome
when planning studies. Additionally, the study suggests the value of incorporating previous
assessments of teacher mediators into the design of research plans.

Authors

• Amota Ataneka, uni of cincinnati

• Fangxing Bai, uni of cincinnati

• Ben Kelcey, University of Cincinnati

• Yanli Xie, Florida University

• Nianbo Dong, University of South Carolina

• Amota Ataneka, Uni of Delaware

• Geoffrey Phelps, ETS