The principle reason for identifying highly gifted children is to help them get a better education than they probably would get otherwise. Tests designed for age- peers are powerless in yielding information to meet this end. One effective method of discovering highly gifted students is by above-level testing. This article focuses on forms of above-level testing with emphasis on the Went search model. Guidelines for above-level testing are offered.

Above-Level Testing
Dr. Leta Stetter Hollingworth first introduced above-level testing in 1916 when she tested "Child E" on the original Stanford-Binet Intelligence Scale (Terman's revision of the Binet-Simon Scale). “The Binet-type scale might be considered the original examination kid table for extensive out-of-level testing" (Stanley, 1990, p. 167). Although Child E was 8 years 4 months at the time of testing, he breezed through all of the tests intended for 12-year-old children. He then proceeded to earn 27 more months of mental age credit at the 14-year, Adult and Superior Adult levels. Given the relative ease with which Child E responded to items on the adult tests, Hollingworth concluded that the test did not have enough ceiling to reflect his full abilities, and, therefore, that the IQ score of 187 was surely an underestimate (Garrison, Burke, & Hollingworth, 1917). She subsequently administered several college-level tests to Child E (and others), including the Thorndike Mental Tests for Freshmen used for entrance to Columbia University and the Army Alpha Test. Hollingworth was convinced of the need for above-level tests and continued to advocate their use with highly gifted children.

Ceiling Effects
As Hollingworth reported, the overriding obstacle in discovering highly gifted children through testing is the ceiling effect, operationally defined as the clustering of scores at the upper limit of the test. Group aptitude and achievements tests and some individual tests are simply too easy for highly gifted students. Ninety-ninth percentile scores on annual tests may please educators and parents, but they do not yield a true picture of the specific functioning of the child. For the highly gifted child, grade-level test scores tell only the percentage of students that performs below the individual but obscure what the child could have achieved had the test included appropriately difficult items. The problem is analogous to that of trying to measure the heights of 12 year old children using a measuring stick that is only 5 feet long. Many children can be measured using the stick, but we cannot differentiate among those who are 5 feet tall and those who are almost 7 feet tall (Stanley, 1990). The solution is to use a longer stick. With testing, the solution is to use an above-level test.

Scholastic Aptitude Test
In 1971, inspired by the work of Leta Hollingworth, Julian Stanley initiated the concept of above-level testing on a national scale through the talent search model (Stanley, 1990). The Scholastic Aptitude Test (SAT) used in the talent searches has become our most tried and true method of discovering highly gifted students at the junior and senior high school levels. In fact, for some highly gifted students trapped in grade-level curriculum, the SAT is the only vehicle that allows them to show their fall range of abilities.

The purpose of the talent search programs is to discover extreme math and verbal abilities among junior high school students who "hit the ceiling" of an in-grade standardized achievement test, that is, those who score at or above the 97th (in some cases, 95th) percentile. The students then take the SAT (normally taken by college-bound juniors and seniors) as an above-level test to help determine levels of ability. The SAT, with its high ceiling, is an effective test for highly gifted students, and plays a uniquely useful role. Even exceptionally gifted fifth and sixth grade students take the SAT in some locations (e.g., The University of Denver). IQ Tests are not effective above middle school level due to ceiling effects, and the value of IQ scores seems to be misunderstood and disparaged. However, the SAT is usually perceived as yielding scores that translate into meaningful, relevant information regarding students' abilities. Math SAT's of 700, for example, mean eligibility for prestigious colleges and imply that the pre-algebra sequence is inappropriate. In addition, the SAT Talent Search program may be well received because it is so huge and pervasive. Each January more than 100,000 seventh and eighth grade students in the United States take the SAT (and recently the ACT), and the program has also expanded internationally. It is an inexpensive way to assess large numbers of students. For these and other reasons, the SAT has become an accepted avenue for discovering older highly gifted students.

Because it is well-respected among educators, the SAT used as an above-level assessment remains a catalyst for educational change for identified students. When teachers are confronted with extremely high SAT scores, many are moved to reexamine the relevancy of their programs for gifted students. Gross (1989) reported that Adrian Seng, a highly gifted Australian boy, scored 760 on the SAT- Mathematics (SAT-M) at the age of 8 years 10 months. Appropriately, Adrian's school responded to this astounding score by designing a highly individualized program of subject acceleration, grade-skipping, curriculum compacting, and enrichment. In short, because school personnel believed Adrian's extreme SAT-M was credible, and reflected his extreme giftedness, Adrian was permitted to develop his immense abilities within a school program.

VanTassel-Baska (1986) showed the appropriateness of using the SAT as an above-level test, and as a diagnostic tool. The following table was adapted from table presented in her article.

The SAT spreads the scores of students who appear to be at the same level, allowing finer discrimination of abilities at the high end. The 2% of students scoring 200-300 are substantially different in abilities and needs from the 1% scoring 710-800 on the SAT-M. We must remember that in-grade testing results tell us that this group of students is homogeneous, scoring at the 99th percentile, while above-level testing gives more specific information regarding the levels of individuals' intellectual talents.

It is not enough, however, to know a students score. The SAT, as with any other above-level test, should be used in order to plan program options that are difficult enough to be challenging and that enhance students' understanding of their own abilities. VanTassel-Baska (1984) discussed program options based on ranges of performance on the SAT. Honors level work, enrichment seminars, and university summer programs were suggested for the 200-520 range. Individualized programming based on diagnostic-prescriptive approaches, early access to Advanced Placement Courses, grade acceleration and early college entrance were suggested for the 500-800 range. A more recent discussion of the curricular implications of talent search scores can be found in Cohn (1991).

In sum, the talent search model embodies important principles used to discover highly gifted students. The SAT testing is accepted by educators and is available to virtually all high achieving students.

Other Above-Level Tests
Unlike the talent searches where testing is used to help prescribe a wide variety of instructional provisions, the Advanced Placement Program (AP) uses a standard instructional provision (college-level coursework) to prepare advanced students for above-level tests. The college-level tests used in AP generally are difficult enough (have high enough ceilings) to yield diagnostic information regarding students' achievement levels in the areas tested. The long-term benefit is that students can earn early college credit on 24 exams in anyone or more of five general disciplines. Although AP courses are often taught in secondary schools, some highly gifted students find particular benefit in enrolling in AP courses at the middle school level or through a correspondence program. For the highly gifted student, success in the AP program provides the flexibility that allows quicker entry into advanced courses in college or universities than is permitted generally.

The College Level Examination Program (CLEP) is a similar program with the unique advantage of obtaining college credits without enrolling in courses or correspondence programs. Students may take exams in five general areas and in 30 specific subjects. As with AP, each college decides which credits and the amount of credits it will award. CLEP is a wonderful way for highly gifted students who might find even the AP courses unchallenging to be tested on their knowledge and to gain some kind of meaningful credit for it.

It is important to recognize the role that these types of above-level tests play in determining appropriate educational plans. For example, Kay Bruce, a highly gifted girt from rural Minnesota, used CLEP to gain credit for many of the general requirements at her college. As a result, her first year at college was filled with the challenge of upper-level college courses. She completed her four-year degree in two years, and, at age 19, was happily contributing to society in her role as a professional social worker.

The International Baccalaureate (IB) program offers advantages similar to AP and CLEP with the unique advantage of its standards of achievement being accepted in higher institutions around the world. Many highly gifted students discover and develop their talents through this program of rigorous work and internationally standardized examinations.

Action
Highly gifted students deserve rich and substantial educations commensurate with their advanced abilities. Locating highly gifted students is the first step on the path to sensible educational planning for them. The following guidelines are offered for using above-level tests to help discover highly gifted students.

Guidelines for Using Above-Level Testing

1. Follow sound general principles of measurement: Use tests with good reliability and validity.



2. Use tests that have sufficient ceilings. The Stanford-Binet Intelligence Scale (Form L-M) is the test of choice for highly gifted students (Silverman & Kearney, 1989). The most commonly accepted above-level test is the Scholastic Aptitude Test: Mathematical and Verbal scales (SAT-M and SAT-V). Other above-level tests include the Raven's Advanced Progressive Matrices (especially useful with many culturally diverse and non-English speaking groups), Advanced placement examinations, the College Level Examination Program (CLEP), and the International Baccalaureate.



3. If tests with sufficient ceilings are unavailable, establish an appropriately high ceiling by using forms of tests intended for older students. The general rule is to retest at appropriate intervals until the student's score falls within two standard deviations of the mean of the test. For moderately gifted children, use a form of the test intended for students one to two years older. For highly gifted students, use a form of the test at a sufficient level of difficulty as to yield diagnostic information.



4. Use aptitude or achievement test scores to help diagnose learning needs. Above-level testing can help discover highly gifted students, and more importantly, should help determine goals for learning.



5. Encourage testing service companies to offer supplements to user's manuals that would assist educators in administering tests and interpreting scores of gifted and highly gifted students (much as the supplements for special education students).



6. Collect qualitative data regarding students' achievements. There have been cases, of phenomenal scores on standardized tests being the sole indicators of some underachieving highly gifted students. However, the use of portfolios, questionnaires, and interviews with parents and significant adults in children's lives have also proven useful in discovering highly gifted students. Through observational information, educators can view a wide variety of achievements and behaviors indicating exceptional ability, when test scores are insufficient.

References

Cohn, S.J. (1991). Talent searches. In N. Colangelo and G.A. Davis (Eds.). Handbook of gifted education (pp. 166-177). Boston: Allyn & Bacon.

Garrison, C.G., Burke, A., & Hollingworth, L.S. (1917) The psychology of a prodigious child. Journal of Applied Psychology, 1, 101-110.

Gross, M.U.M. (1989). Children of exceptional intellectual potential: Their origin and development. Unpublished dissertation. Purdue University, W. Lafayette, IN.

Silverman, L.K. & Kearney, K. (1989). Parents of the extraordinarily gifted. Advanced Development, 1, 41-56.

Stanley, J.C. (1990). Leta Hollingworth's contributions to above-level testing of the gifted. Roeper Review, 13, 166-171.

Van Tassel-Baska, J. (1984). The talent search as an identification model. Gifted Child Quarterly, 28, 172-176.

Van Tassel-Baska, J. (1986). The use of aptitude tests for identifying the gifted: The talent search concept. Roeper Review, 8, 185-189.

Permission Statement

Permission to reprint this article has been granted to The Davidson Institute for Talent Development by the author.

This article is provided as a service of the Davidson Institute for Talent Development, a 501(c)3 nonprofit dedicated to supporting profoundly gifted young people under 18. To learn more about the Davidson Institute’s programs, please visit www.DavidsonGifted.org.