
Item Response Theory (IRT) scales are available for each subject assessed beginning with the 1984 assessment. Content of the subject area scales is specified in the subject area frameworks, which have been developed by the National Assessment Governing Board.
While the arts framework specifies that students' knowledge and skills be measured in four arts disciplines, grade 8 students in 2008 were assessed in music and visual arts only, due to budget constraints and the small percentage of schools with dance and theater programs. Each area had one IRT scale, and results were reported independently for each area. No overall “arts” result was reported.
For grades 4, 8, and 12, even though there are three framework-defined "purposes" of civics, one IRT scale—as established in the civics framework in 1998—has been used in all main civics assessments.
For grade 12, three IRT scales, as defined by the economics framework, were established in 2006. A composite scale was formed at each grade by taking a weighted sum of the economics content-dimension scales:
For grades 4, 8, and 12, three IRT scales, as defined by the geography framework, were established in 1994. A composite scale was formed at each grade by taking a weighted sum of the geography content-dimension scales:
For grades 4, 8, and 12, five IRT scales, as defined by the mathematics framework, were established in 1990. A composite scale was formed at each grade by taking a weighted sum of the mathematics content-strand scales:
Starting in 2005 the grade 12 mathematics assessment contained four IRT scales. A composite scale was formed by taking a weighted sum of the mathematics content-strand scales:
For grade 4, two IRT scales, as defined in the reading framework, were established in 1992. At that time, three IRT scales were established for grades 8 and 12. A composite scale was formed at each grade by taking a weighted sum of the purpose-for-reading scales:
For grades 4, 8, and 12, three IRT scales, as defined in the science framework, were established in 1996. A composite scale was formed at each grade by taking a weighted sum of the field-of-science scales:
For grades 4, 8, and 12, four IRT scales, as defined in the U.S. history framework, were established in 1994. A composite scale was formed at each grade by taking a weighted sum of the historical themes scales:
For grades 4, 8, and 12, even though there are three framework-defined "purposes" of writing, one IRT scale—as established in the writing framework in 1998—has been used in all main writing assessments since (i.e., 1998, 2000, 2002, and 2007).
Longitudinal studies for ages 9, 13, and 17 were conducted in 2004 and 2008 to track trends for mathematics since 1973 and for reading since 1971. For each long-term trend subject (i.e., mathematics and reading), a univariate scale was used in assessing each age group.