Estimating Student Achievement at the Topic Level in TIMSS Using IRT-Based Domain Scoring


Title:

 Estimating Student Achievement at the Topic Level in TIMSS Using IRT-Based Domain Scoring


Description:

Student achievement data from large-scale assessments have traditionally only been analyzed at the overall subject (for example, mathematics) or content area levels (for example, geometric thinking). This report introduces a method to analyze large-scale student achievement data at the topic level (for example, using fractions), using data from the Trends in International Mathematics and Science Study (TIMSS) fourth-grade mathematics assessment. Obtaining topic-level scores from large-scale assessments, such as TIMSS, allows for the identification of specific areas of strength and weakness in student performance and how those may be related to curricula. The introduced method has the additional benefits of straightforward implementation and intuitive interpretation.


Online Availability:



Need Help Viewing PDF files?


Cover Date:

February 2020


Web Release:

February 25, 2020


Publication #:

NCES 2020038


Center/Program:

NCES


Authors:

Stephen Provasnik, Enis Dogan, Ebru Erberber, and Xiaying Zheng


Type of Product:


Technical/Methodological Report


Survey/Program Areas:


Trends in International Mathematics and Science Study (TIMSS)


Keywords:



Questions:



For questions about the content of this Technical/Methodological Report, please contact:
Stephen Provasnik.


 


 





Source link