The identification of gifted and talented learners has persisted in being a problem for educators at the local level, even though many guides and books have addressed the issue over a number of years. There are perhaps many valid reasons for this situation:
Coupled with these general problems have been key research directions in the field that have raised uncomfortable questions about how tests are used to identify the gifted.
Many investigators in the field of gifted education have decried the use and abuse intelligence tests (Renzulli, 1978; Sternberg, 1983). Others have found the use of a standardized testing procedure inappropriate for identifying the most talented (Alvino, 1979; Bruch, 1971; Torrance, 1970), particularly for students from minority backgrounds. And most school districts continue to adopt identification procedures that attempt to balance testing with more observational sources like teacher recommendations and peer or parent inventories. Yet no studies have refuted Martinson's Claim (1974) that the individual Stanford Binet is still the single best measure to identify gifted children, and Richert (1982) reports it to be the best individual intelligence test available for identifying general intellectual ability.
One important development regarding the identification dilemma was proposed by Stanley. Keating, and Fox (1974). Their simple but elegant idea, first suggested by Hollingworth, was to administer a more difficult test, one normed on older children, to a younger group. Specifically, he recommended using the Scholastic Aptitude Test (SAT) to find precocious seventh grade students. The efficiency of this approach is no legendary. Close to 100,000 students are now tested annually through this procedure nationally and internationally.
The talent search model employs a two-step testing protocol. At step one, it seeks to find all students who score at the 95th or 97th percentile on their in-grade standardized achievement test battery, either in mathematical or verbal areas. Individual talent searches employ different cutoffs at these stages of eligibility. At step two, it discriminates within this population by administering the Scholastic Aptitude Test (SAT) typically used at senior high school level to identify students who will, do well in college. Because there is a dispersion of scores on the SAT such that there is no ceiling effect, educators can better discern the potential of these junior high age students at this critical stage of development -I in key academic aptitude areas. Furthermore, because such younger students usually have not had advanced coursework in mathematics or the verbal arts, the scores reflected are mere representative of aptitude rather than achievement thus countering the charge frequently made about the SAT when used with older populations, that it measures achievement rather than aptitude.
Yet there are detractors within the field of gifted education regarding appropriateness of using this sophisticated measure with junior high age populations. Their arguments may be summarized as follows:
1. The SAT creates an unhealthy testing situation for younger students. They will become anxious and traumatized just by the testing experience itself.Generally speaking, the over 200,000 students of junior high age who have been exposed to the SAT over the last eleven years have not experienced test anxiety as a result of taking it. A small study (Feldhusen, 1983) found mild anxiety among junior high age test takers: yet both test literature and common sense would argue that some anxiety is positive in a testing situation. However, all talent searches recommend that parents and/ or school personnel orient students to the test. Toward that end, students review a practice test provided to them by the talent search in order to be familiar with the format as well as the nature of content terms that they will find on the test. In addition, school personnel and parents attempt to stress the following issues with students:
2. The SAT should not be used with such large numbers of students since the only benefit accrues to students who score at extremely high levels in a given area on the test.The SAT performs admirably in the task of discriminating within a gifted population. In this sense, it functions like an IQ test at earlier ages. The assessment information accrued should act as a basis for educational planning for all students to whom it is administered. And school districts should and in many cases do actively utilize the data for identification and placement in local programs as well as see it as a vehicle for universities and other agencies which offer special programs for talent search populations.
One of the advantages of testing large numbers is to have the potential for effecting change at all levels of the educational enterprise, and to systematically recognize high level academic talent regardless of how a student may score at the second level of the testing protocol. Local, regional and state awards ceremonies have proliferated for talent search students based only on participation in the testing experience.
3. The SAT discriminates against disadvantaged students in its singular focus on achievement and academic aptitude.A recent incidence study in the midwest has found that 15.2% of the talent search students are disadvantaged. When using an economic definition of that term (Van Tassel-Baska and Chepko-Sade in press). Furthermore, 11% of talent search students in the midwest are from minority cultures. Of that number only 10 percent fewer disadvantaged students than their advantaged counterparts score at qualifying levels for programming opportunities. Based on these data, it seems fair to claim that the SAT aids us in finding academically able disadvantaged gifted students and creates an important talent pool of individuals who are in need of scholarship assistance to gain educational advantages typically reserved for their economically advantaged counterparts.
The challenge in this issue is to systematically encourage and prepare disadvantaged gifted students to participate fully in the talent search experience. For some, economical and family circumstances may otherwise limit their involvement. But the issue of recruitment and follow up of such students is not resolved by attacking the use of standardized tests to help find them.
What important information does aptitude testing on the SAT provide educators as they institute local identification policies? The principles embodied in the talent search identification model have only recently begun to be employed within school districts to identify those students most in need of a differentiated program at the -local level. Yet these principles translate very well to local program identification issues. A discussion of three central principles embodied in the talent search identification model will be discussed and illustrated by data collected through the Midwest Talent Search Project at Northwestern University.
The Principle of Off-Level TestingBecause gifted children always tend to score toward the top end of an in-grade standardized test, it is appropriate to attempt to discriminate true potential based on scores from such tests. It also is unjustifiable to establish an arbitrary cutoff within the narrow band at the top of the test. True potential for specific academic work in mathematics and verbal areas can only be discerned by administering an off level test, standardized on older populations. Many local school districts ignore this principle by arbitrarily using cutoff points within a narrow band of ability on in-grade measures. Consequently, many able learners are being overlooked in the process.
A recent analysis conducted by Midwest Talent Search illustrates this principle quite well. We analyzed outcome scores on the SAT for the subjects in our 1985 talent search who sat for the SATs compared with their reported percentile scores on in-grade standardized achievement tests. The cutoff score for participation in -Midwest Talent Search is the 95th percentile, compared to the 97th percentile for the Johns Hopkins University and Duke University projects. We were interested in ascertaining the extent to which students in those lower percentiles (95-98) were scoring at a level on the SAT sufficient to qualify for special programs. Table I demonstrates that 1.228 (48%) students of all 2,574 students scoring in the 95th and 96th percentile were capable of scoring 410 or higher on the mathematical section of the test (The arbitrary level of 410 was used due to program eligibility in many summer opportunities in the Midwest.) Two-hundred seventy-one students (15%) scored above 510 on SAT-M, a level that further qualifies them for participation in programs with more stringent standards for admission Northwestern University, however, uses a 500 SAT-M cutoff for its math and science programs. However, in the upper ranges of the test (SAT M 610) only eight students scoring at the 95th and 96th percentile were found.
In the verbal section of the test, 28.8% (N =734) of students at the 95th and 96th percentile (N=2,548) scored SAT-V > 410, as shown by Table 2. The score of 4l0 on the verbal portion of the test is a sufficient qualifying score for many summer programs run in the Midwest for MTS students. However, again as in the case of the math scores, only 14 students in these percentile levels scored SAT –V>=610.
© 1986 Permission to reprint this article granted by the author Joyce Van Tassel-Baska
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 18 and under. To learn more about the Davidson Institute’s programs, please visit www.DavidsonGifted.org.
The appearance of any information in the Davidson Institute's Database does not imply an endorsement by, or any affiliation with, the Davidson Institute. All information presented is for informational purposes only and is solely the opinion of and the responsibility of the author. Although reasonable effort is made to present accurate information, the Davidson Institute makes no guarantees of any kind, including as to accuracy or completeness. Use of such information is at the sole risk of the reader.