方法对比

	Conditional Standard Error of Measurement	Standardized Test Analysis
领域	Education	Education
方法族	Latent structure	Latent structure
起源年份	1980	2014
提出者	Test theory (Lord; Feldt; codified in the Standards)	Educational measurement profession (AERA/APA/NCME Standards; Lord; Cronbach)
类型	Measurement-error estimate that varies with the examinee's score or ability level	Psychometric evaluation of items, reliability, validity, and fairness of standardized tests
开创性文献	Lord, F. M. (1980). Applications of Item Response Theory to Practical Testing Problems. Lawrence Erlbaum Associates. ISBN: 9780898590067	American Educational Research Association, American Psychological Association, & National Council on Measurement in Education. (2014). Standards for Educational and Psychological Testing. AERA. ISBN: 9780935302356
别名	CSEM, Conditional SEM, Score-Level Measurement Error, Conditional Standard Error	Standardized Testing Analysis, Test Score Analysis, Item and Test Analysis, Educational Test Psychometrics
相关	3	3
摘要	The conditional standard error of measurement (CSEM) describes how much measurement error a test score carries at each point along the score scale, rather than as a single average. A test typically measures more precisely in some score ranges than others — often best near the middle and worst at the extremes — and the CSEM captures that variation. Recognized in test theory by Lord and required by the Standards for Educational and Psychological Testing, it is essential for honest score reporting, especially near cut scores where classification decisions are made.	Standardized test analysis is the body of psychometric methods used to evaluate and score standardized educational tests: analyzing how items perform, estimating reliability and the standard error of measurement, scaling scores via classical or item response theory, and assembling validity and fairness evidence. Governed by the professional Standards for Educational and Psychological Testing and rooted in test theory synthesized by Lord and others, it is the disciplined work that turns a set of test questions into defensible scores carrying meaning, precision, and fairness.