Grant Open

Design Comparable Effect Sizes for Single Case Designs

NCER

Program: Statistical and Research Methodology in Education

Program topic(s): Core

Award amount: $899,721

Principal investigator: Larry Hedges

Awardee:

Northwestern University

Year: 2022

Project type:

Methodological Innovation

Award number: R305D220052

Purpose

An important function of effect size measures is to represent the results of studies in a manner that is as comparable as possible. Single-case designs are more difficult to compare across studies, given the typically low number of participants involved in such studies. The purpose of this project is to develop design comparable effect size measures for alternating treatments designs and for changing criterion designs.

Project Activities

The research team will first derive the approximate sampling distributions of the design comparable effect sizes. They will then evaluate the accuracy of the distributions using Monte Carlo simulations of three different estimators of the effect size each for alternating treatments designs and for changing criterion designs.

People and institutions involved

IES program contact(s)

Charles Laurin

Education Research Analyst

NCER

Products and publications

Products: After any necessary mathematical adjustments and reevaluation, the researchers will develop software for calculating the effect size estimates; the software will be available in R and as a Shiny app online. Other products include manuscripts in methodological journals to support additional development of design comparable effect sizes and manuscripts in applied journals to facilitate the use of the effect size estimates by applied researchers.

Publications:

Hedges, L.V., Shadish, W.R. & Natesan Batley, P. (2023). Power analysis for single-case designs: Computations for (AB)k designs. Behavior Research Methods, 55, 3494-3503.