Data from large-scale international studies reflects the nested structure of education systems and is, therefore, very well suited for hierarchical linear modeling (HLM). However, because this data comes from complex cluster samples, there are methodological aspects that a researcher needs to understand when doing HLM, e.g., the need for using sampling weights and multiple achievement values for parameter estimation. This course teaches participants how to do HLM with data from these studies, including PIRLS, TIMSS, and PISA.
Find specific work or narrow your results by type, topic, program, project, or service by selecting your criteria from the choices at right.
31 Dec 2005
Through our assessment services, AIR works together with countries and partner organizations to develop appropriate expectations of student performance, create instructional materials aligned to learning standards, measure student learning, and apply the data to help teachers teach more effectively.