Ages: 5-0 through 14-11
Testing Time: 20 minutes each
Administration: Individual or group

The SAGES-­2 is helpful in identifying gifted students in kindergarten through eighth grade. Its three subtests sample aspects of two of the most commonly used areas for identifying gifted students: aptitude and achievement. Aptitude is measured via the Reasoning subtest: The student is asked to solve analogical problems by identifying relationships among pictures and figures. The other two subtests assess achievement. On one of these subtests the child answers questions about language arts and social studies, and on the other, about mathematics and science. The student selects answers from a series of pictures, symbols, or words. The subtests can be used to examine the relationships between aptitude and achievement. The SAGES-­2 can be used with students ranging in age from 5-0 to 14-11. Although they are untimed, each subtest requires approximately 20 minutes to administer. All of the SAGES-­2 subtests can be administered in small groups or individually.

The SAGES-­2 was normed on two large samples tested in 1998 and 1999. Sample One (normal sample) consisted of 3,023 students who were in heterogeneous classrooms, and Sample Two (gifted sample) consisted of 2,290 students who were identified as gifted by their local school districts. The demographic characteristics of both samples were matched to those of the United States according to the 1997 census. The normal normative sample was stratified on the basis of age, gender, race, ethnic group membership, and geographic location. Standard scores and percentile ranks are provided for both samples. 

Several important improvements to the technical characteristics of the initial SAGES were made. First, test­retest studies were added. Second, each item on the test was evaluated using both classical item analyses and item response theories to choose "good" statistical items. Third, differential item functioning analyses were performed on three dichotomous groups in order to find and eliminate adversely biased items. Finally, several new validity studies were conducted with special attention devoted to demonstrating that the test proves valid for a wide variety of subgroups as well as for a general population. 

The reliability coefficients for the test are high, ranging from .77 to .95. Ninety-seven percent of these reach or exceed .80, and seventy-four percent reach or exceed .90. Test­retest studies show that the SAGES-­2 is stable over time. 

The potential bias of every item on the test on the basis of gender and ethnic group was studied. In all, 44 items were eliminated from the final version. 

Extensive validity data are reported as well, documenting the test's relationship to the WISC-III, OLSAT, Stanford Achievement Test, and Gifted and Talented Evaluation Scale, and its efficiency in discriminating groups appropriately. The SAGES-­2 was built to identify students as gifted. We found that at most raw score values, the gifted sample scored 1 standard deviation higher than the normal sample. In addition, the test clearly measures high-ability children with less error.