Applying a Thurstonian, Two-Stage Method in the Standardized Assessment of Writing

ABSTRACT

National standardized assessment programs have increasingly included extended written performances, amplifying the need for reliable, valid, and efficient methods of assessment. This article examines a two-stage method using comparative judgments and calibrated exemplars as a complement and alternative to existing methods of assessing writing. Written performances were taken from Australia’s National Assessment Program—Literacy and Numeracy, which included both narrative and persuasive performances from students aged 8 to 15. In Stage 1, assessors performed comparative judgments on 160 performances to form a scale of 36 calibrated exemplars. These comparative judgments showed a very high level of reliability and concurrent validity. In Stage 2, assessors scored 2,380 new performances by matching them to the most similar calibrated exemplar. These matching judgments showed a generally high level of reliability and concurrent validity and were reasonably efficient after a familiarization period. Further research is suggested to enhance Stage 2 by simplifying the exemplar scale and scaffolding it with detailed descriptors. Overall, the findings support the use of the method in standardized writing assessment and its application to various learning areas.

Notes

¹ The lowest score of 50 was not intended to indicate a “pass” performance, but rather as a more intuitive lower bound for assessors than the negative values of the logit scale.