In order to plan group-level intervention studies, such as school- or classroom-based interventions, researchers must understand the proportion of the variation in a key outcome (such as a state test) that occurs within a school or classroom versus the variation that occurs between schools or classrooms. This is typically expressed by calculating or estimating an intra-class correlation coefficient (ICC). In planning a particular research design, investigators need good estimates of the relevant ICCs to perform power analyses, determine the needed sample size, and create a study budget. This study proposes to determine the ICC values for statewide achievement tests in various subjects and grade levels, and to make these values widely available for researchers who are planning to use state assessment data as outcome measures for experimental and quasi-experimental designs.
In order to plan group-level intervention studies, such as school- or classroom-based interventions, researchers must understand the proportion of the variation in a key outcome (such as a state test) that occurs within a school or classroom versus the variation that occurs between schools or classrooms.
