Gifted Child Quarterly
National Association for Gifted Children (NAGC)
This article by Joyce VanTassel-Baska describes the advantages of the "Talent Search Identification Model" for gifted student identification. The widely used model employs standardized achievement tests to identify junior high students. Recommendations are offered for students scoring within specified ranges on the Scholastic Aptitude Test.
Over the last six decades, the identification of a phenomenon called "giftedness" has preoccupied some of the best minds in educational psychology. Out of this preoccupation have emerged several operative models which have come to be accepted by educators working with gifted students within the context of schools. Still, the most prevalent model appears to be one based upon scores from intelligence tests (Alvino, McDonnel, Richert, 1981). Use of the individual intelligence testing model has been widely supported by early research, (Terman, 1925) and by knowledgeable practitioners (Martinson, 1974). Regulations in several states require the use of the IQ model with a specific minimum score, usually 132 or above. With the expanded definition of giftedness emanating from the federal level in the seventies (Marland report, 1972), a second type of identification model became popular. This model might be called the "multiple criteria" model because it focuses on the use of at least three criteria in the identification process, and seeks to open up the identification of gifted students into categories other than intellectual. Use of this model has also been very prevalent across the country (Gallagher, 1975). More recent models have focused on nonbiased assessment issues in the identification of minority students (Meeker, 1978; Mercer, & Lewis, 1978; Baldwin, 1978; Frasier, 1983). The revolving door identification model (Renzulli, Reis, & Smith, 1980) has recently enjoyed popularity as it seeks to create a larger talent pool of students for program consideration. Also, with the advent of more early childhood programs for the gifted, identification models that place strong emphasis on the role of parents in the process have gained acceptance (Roedell, Jackson, & Robinson, 1980; Ehrlich, 1982; Karnes, & Bertchi, 1979).
Of all the models for identification currently being used, however, only the Talent Search Identification Model offers a standardized approach, on a national basis, that systematically addresses the procedures of screening, verification, and placement, seen to be crucial to the development of an identification technology (Foster, 1979). It also is the only model in which students and parents have equal access to testing regardless of local program standards, and where the most cost-efficient approach to testing is employed.
The state of Illinois, having utilized the model on a statewide basis since 1978, credits the talent search with having the greatest impact on program development and articulation of any single effort in gifted education (Stanley, & Benbow, 1983). Similarly, over the last two years, the state of Indiana has experienced direct program growth as a result of the talent search model (VanTassel-Baska & Prentice, in press). Efficacy of the model in respect to the appropriateness of the instrumentation has also been aptly demonstrated (Stanley, George, & Solano, 1977; Stanley, & Benbow, 1981, 1983). In addition, case study follow-up work has shown the predictive validity of the model for finding students who can perform well in advanced academic settings (Stanley, Keating, & Fox, 1974; Keating, 19Th Benbow, & Stanley, 1983).
In keeping with its research promise, the Talent Search Identification Model has now expanded to national and international dimensions. The model has stood the test of time, being utilized over a 12-year period, experience wide-spread usage in all parts of this country and abroad. Five talent search programs utilizing the model operation across the United States at Johns Hopkins, Duke, Arizona State, Denver, and Northwestern Universities. Over 80,000 students are tested annually through these projects. The identification protocol itself is very simple and, perhaps because of that simplicity, very effective. Basically, the model (developed by Julian Stanley and his colleagues at the Johns Hopkins University) utilizes a two-step protocol.
- It identifies students of junior-high age who have already scored at the 95th percentile (the Johns Hopkins and Duke talent search use 97th percentile) or higher on an ingrade standardized achievement test in their homeschool setting to determine eligibility.
- It uses the Scholastic Aptitude Test (SAT) as a second-level test to determine level of mathematical and verbal ability.
The model also provides insight into three key issues in identifying gifted youth: (a) establishing a talent pool (b) combating "ceiling effect" on tests, thus providing good score discrimination; and (c) providing effective data for program planning at the school district, university, and state levels.
Establishing a Talent Pool
Some in the field of the education of the gifted would argue that too few criteria are used for selecting students according to the talent search model. Many state regulations, for example, mandate the use of a minimum of three criteria for selecting students who may be deemed gifted and talented. It seems important to point out that the talent search model does employ a minimum of three criteria for selection, in that there exists "a hidden step" represented by student interest, motivation, and persistence. While it is true that students must be at least at the 95th percentile or higher on an ingrade, standardized achievement test to qualify for the SAT testing, they also must possess the necessary motivation to take the responsibility for signing up for the testing, to follow through and devote a Saturday of their own time to participate in the testing, and then follow up from there in terms of urging schools to provide special programs and services. So, in addition to the testing protocol of achievement coupled with aptitude testing, the talent search also uses this third criterion of student motivation and interest. That does not differ significantly in kind from efforts that are frequently employed in schools where student interest inventories or student nomination forms are included as part of the identification system.
The use of second-level testing in the talent search model employs an important mechanism for selection purposes. In recent years school districts have focused on the idea of a "talent pool" as a way of describing the larger group from whom students would be selected for programming purposes. In the talent search model, the size of the talent pool in any given year comprises a standardized-tested group in the upper 5% nationally which may translate to as many as 20% within a local school district setting or as few as 1%, depending on the local situation. And that pool, in essence, is narrowed significantly and systematically by sophisticated instrumentation at the second level. The appropriateness of that instrumentation has been shown over the last 12 years of research (Stanley, Keating, & Fox, 1974; Stanley, & Benbow, 1983). Thus, the use of the SAT would seem to offer not only the most predictive instrument for academic work for students ready to go on to an advanced level, but also provides the best discriminator yet found for discerning degrees of giftedness in terms of academic potential within the larger gifted and talented population.
Score Discrimination Among Gifted Populations
The off-level or out-of-level testing phenomenon the talent search model embodies is an important principle in the identification of the gifted. It can be utilized in order to discriminate well among gifted students, particularly for offering diverse program options on a mild-to-intensive intervention system. It also is important to avoid "ceiling effect," defined as the tendency for gifted students to cluster together at the top level of a test (Keating, 1975). While educators may know that there are many gifted students or whom special programming is desirable, prior to the talent search identification model, they have not known the degree of giftedness many of these students possess. Consequently, the nature of the program alternatives offered has not necessarily been appropriate or well-matched to tested ability. The talent search model does provide that kind of information for direct and systematic follow-up intervention. The following recommendations show the nature of program intervention that may be desirable, given the spread of scores on the Scholastic Aptitude Test for Talent Search participants. These recommendations are also portrayed graphically in Figure 1.*
Specific Recommendations for Students Scoring 200-390 on SAT-V and SAT-M
Even though a student may not score at a level considered appropriate for placement in a special, fast-paced class or program, there are still many opportunities that can and should be provided for these students who take the test. It is important to remember that each student who participates in a Talent Search project would probably be identified as gifted and talented within his own school district and, therefore, should be entitled to the following kinds of special opportunities:
High-level coursework within the middle school or junior high school setting in the area(s) of strength.
- Follow-up academic counseling to examine future course-taking and academic directions.
- Special enrichment seminars in areas of interest that may be held in addition to school work and on Saturdays.
For example, the Chicago Public Schools offer an enrichment series of eight lectures to all search participants and their parents, held on Saturday mornings at the Museum of Science and Industry, Another example would be the special courses which are offered at Purdue University and Northwestern University on Saturdays during the academic year.
Specific Recommendations for Students Scoring 400-520 on SAT-V and/or
Students scoring at these levels on the SAT can profit from special accelerated coursework at the seventh- and/or eighth-grade level in each respective area of aptitude. For example, a seventh grader scoring SAT-M ≥ 400 can handle Algebra I. In the verbal area, an SAT-V ≥ 400 for a seventh grader would indicate ability to handle an intensive writing skills program, a special critical reading program, or a high school foreign language offering (e.g., Latin I) in less time than even other able students of their age. Thus, the following recommendations are made:
- Offer fast-paced, advanced coursework during the academic year in a student's tested area of strength.
- Provide follow-up academic counseling to encourage student involvement in advanced coursework and to help students set educational goals.
- Encourage these students to attend a summer program on a university campus in an academic area of strength and interest.
Specific Recommendations for Students
Scoring 530-650 on SAT-V and/or
Students scoring at these very high levels on the test should be encouraged to participate in as many academically appropriate educational alternatives as possible. These include but are not limited to the following:
Develop an individualized program of study in the home-school setting that will allow these students to "test out" of content already mastered in mathematics and/or verbal areas and to continue to progress in the content area at a rate commensurate with their proficiency. A diagnostic-prescriptive (D - P) teaching approach is a necessary component of such a program.
- Encourage student participation in one of the university programs that employ a fast-paced model. Opportunities for interaction with other highly gifted students, intensive study with high-quality faculty in a compressed time frame, and opportunities for higher-level proficiency are important benefits.
- Provide academic counseling, especially on early access to Advanced Placement (AP) and the possibility for grade acceleration in the span of years from grades 7-12.
Specific Recommendations for Students Scoring 650-800 on SAT-V and/or
Students scoring at the highest range on the test are operating at levels better than 80% of college-bound seniors on the same test and, therefore, need to have additional opportunities and a more intensive program. Recommendations for this group would include all of those listed for the preceding range (530-650 SAT-M or V) and the following additional ones:
Investigate early admission opportunities and/or advanced standing at high quality institutions for these students now. Early planning for the appropriate college that may offer scholarships to outstanding students needs to occur early.
- Establish a mentorship or tutorial for such students pairing them with outstanding adults in the community who share the ability and interest of such students.
- Provide students with easy access to career information that provides data on high-level professional careers focusing on original research.
Figure 1* summarizes recommended program opportunities for all students who participate in a talent search program. Cooperative planning across school districts with universities and other regional entities, is desirable in meeting the needs of many of these students effectively.
The Talent Search as a Diagnostic Tool
for Program Planning
Frequently the testing phase of the program has been perceived as being the major focus of talent search projects and the major purpose, to merely identify students who are very precocious in the mathematics area or in the verb area. In reality, the talent search identification model provides excellent diagnostic information for purposes of placement and program in the two critical areas of mathematical reasoning ability and verbal reasoning ability. The five universities utilizing the model have all systematically employed the information from the SAT for specific diagnostic -prescriptive purposes in summer programs, correspondence work, and commuter programs that specifically address the academic level of students as discerned by their score ranges on the SAT instrument. For example the Northwestern model is shown below in terms of score ranges and program options available:
|SAT-M ≥ 500|
|SAT-V ≥ 430
In addition to providing good assessment information the most highly gifted students who participate in the search for purposes of university intervention, the talent search identification model also provides excellent assessment information for schools and parents, as they try to provide appropriate program strategies at the local level. The score information can be useful in terms of forming advanced classes at the junior high level. It can be useful for school districts considering the acceleration of a gifted student, or set of students in a content area, or grade acceleration in all areas of the curriculum. It can be useful in terms of working out a continuous progress program for gifted students, since it provides more precise information on the level of performance capacity than other tests given by the school.
Many school districts have also found the information useful for purposes of honoring their own students at an awards assembly given in the spring of the year or creating an honors day specifically for the academically able. Because the information comes from a tool that is secure and highly credible for selecting academically able students or college admission purposes, the drama associated with younger students scoring very high on the SAT instrument s perceived as a very noteworthy accomplishment. Consequently, school districts have become more interested in honoring academic excellence as a result of the use of the talent search identification model.
A third area where local programs have benefited from the diagnostic information is in the development of a multilevel service delivery network wherein local, regional, and state organizations work to provide an array of programs and services for the academically able. Let us cite an example of how such a system works. Student A scored 700 on the mathematics portion of the SAT. He is a seventh-grade student in a local district that is willing to shove him into high school honors coursework but unable to work it out from a scheduling perspective. The district then turns to other sources in the community that might provide appropriate mathematics programming. There is junior college nearby where Student A can take high level mathematics in the evening. There is also a retired university professor in the community from whom he can receive individual tutorial help in precalculus. Or he can take a correspondence course from a university, such as Northwestern. Each option represents a different level of program opportunity that can supplement the efforts of the local school. Over time, a resource bank of options for their students at similar levels of ability may be developed and tapped as necessary. Thus, a cooperative network of program services can be available to assist the academically advanced in all geographic areas.
It might be useful to point out some specific reasons why the talent search identification model has become popular and has been more widely adopted than any other identification system. The following points might provide some insight:
The talent search identification model provides a precise, focused and appropriate way of matching the tested ability of a student to a specific curriculum intervention strategy, unlike other identification protocols that focus on the identification of general intellectual functioning.
- Because of its effectiveness in separating those most highly gifted students from those moderately and mildly gifted, it has been a catalyst for curricular change in many school districts. Whereas before school districts could merely say they had a group of students who were in the upper 5% on their standardized achievement tests, they now are forced to recognize the fact that they have students who clearly are operating four, five, and six years beyond their age group in terms of ability. They, therefore, feel much more pressed to provide an appropriate program for them.
- It has been a significant catalyst for bringing about a "change of attitude" about acceleration and has led to the implementation of acceleration policies and practices at both the local and state levels in ways that had not occurred prior to the talent search model being used. Even though the research on acceleration has been as definitive as any educational research could be in terms of the lack of harm as well as the positive benefits that can accrue for students who have had the experience, acceleration has not been widely practiced. The talent search model has clearly shown and demonstrated, on a case study basis, that students have profited enormously from exposure to accelerated programs and coursework. Consequently, districts have begun to give ground on this crucial issue and, specifically, have established policies and procedures to allow for acceleration of academically able students. The state of Indiana, for example, has established a waiver system that takes into account the flexibility needs of individual students as they move through their school program.
- The talent search focuses much more sharply than most identification protocols on self-election or the volunteerism principle. The commitment to talent search and to follow-up procedures must be made by students and parents in order for the identification to occur. It is this volunteerism principle that has been an important aspect in making the model so adaptable to a variety of school and geographic settings.
- The talent search model discerns and rewards reasoning ability at high levels, exclusive of content knowledge, per Se. Too frequently the nature of reward systems in schools has focused on those students who have performed the best in terms of in-class content knowledge, without any recognition being extended to those students who have high potential and high reasoning ability but who, for a whole array of reasons, have not necessarily been the highest performers in the school. Therefore, the talent search is finding and recognizing some of those students who are under-achievers, at least within the context of their local school districts.
- The talent search model cannot be under-estimated as a public relations vehicle for school districts either. It calls attention to their high academic achievers, leading to local honors days, initiation of special programs, board commitment, parent advocacy, and student esteem. The talent search model, in this sense, provides a more concrete way of addressing the needs of the academically talented than ever before in the history of the gifted movement, in that it allows for participation at all levels. There is commitment on the part of educators, parents, and students to a common enterprise in furthering the academic advancement of those students for whom it is deemed appropriate.
In conclusion then, the talent search model carefully researched by Julian Stanley and others at The Johns Hopkins University, now implemented in all fifty states and internationally recognized, represents a systematic way of identifying high level academic talent at a particular point in time in a student's life. It attempts to provide appropriate strategies and program intervention for that talent from the point of identification right on through graduate school.
Furthermore, the talent search model embodies some of the most basic and important principles of good identification of the gifted which should be adhered to for insuring fair and appropriate placement in programs. These principles include:
An effective match between identification protocol and the program planned.
- The establishment of the predictive validity of identification procedures based on student success in a planned program.
- Choice of valid and reliable instrumentation.
- Use of off-level or out-of-level testing procedures to avoid ceiling effect and to more precisely assess high levels of ability.
- Use of a two-step procedure that incorporates effective screening prior to final testing.
- Planned or recommended opportunities for all students who participate in the search screening.
- Specific assessment data that can be used in educational planning with parents, educators, and students themselves.
The talent search identification model, then, is not only consonant with solid principles of identification of the gifted, but also provides an exemplum for educators working in the field to find equally effective models for use with younger populations and those with talents other than academic.
* Please see original for Figure 1. Ranges of Performance on the SAT.
Alvino, J., McDonnel, R.C., & Richert, S. National survey of identification
practices in gifted and talented education. Exceptional Children, 1981, 48 (2), 124-132.
Benbow, C. P., & Stanley, J. C. Academic precocity: Aspects of its
development. Baltimore: The Johns Hopkins University Press, 1983.
Ehrlich, V. Gifted children: A guide for parents and teachers. Englewood
Cliffs, NJ: Prentice-Hall, 1982.
Foster, W. The unfinished task: An overview of procedures used to identify gifted and talented youth. In N. Colangelo, & R. Zaffrann, (Eds.), New voices in counseling the gifted. Dubuque, IA: Kendall/Hunt, 1979.
Fox, L. H. Identification of the academically gifted. American Psychologist,
1981, 36 (10), 1103-1111.
Frasier, M. A non-biased assessment model for the gifted minority student,
a paper presented at the College Board National Forum, Dallas, Texas October, 1983.
Gallagher, J. J. Teaching the gifted child (2nd ed.). Boston: Allyn & Bacon
George, W. D., Cohn, S. J., & Stanley, J. C. (Eds.). Educating the gifted
Acceleration and enrichment. Baltimore: The Johns Hopkins University Press, 1979.
Karnes, M., & Bertchi, J. Identifying and educating gifted/talented non
handicapped preschoolers. Teaching Exceptional Children, 1978, 10, 114 -118.
Keating, D. P. Testing those in the top percentiles. Exceptional Children.
1975, 41(6), 435-6.
Keating, D. P. Intellectual talent: Research and development. Baltimore
The Johns Hopkins University Press, 1976.
Marland, S. P. Education of the gifted and talented. Report to the Congress
of the United States by the U.S. Commission of Education. Washington DC: U.S. Government Printing Office, 1972.
Martinson, R.A. The identification of the gifted and talented.
Ventura, CA Ventura County Superintendent of Schools, 1974.
Meeker, M. Nondiscriminatory testing procedures to assess giftedness in Black, Chicano, Navajo and Anglo children. In A. Y. Baldwin, G.H Gear, & L. J. Lucito (Eds.), Educational planning for the gifted
Overcoming cultural, geographic, and socioeconomic barriers. Reston
VA:Council for Exceptional Children, 1978.
Mercer, T. B., & Lewis, J. G. Using the system of multicultural assessment
(SOMPA) to identify the gifted minority child. In A. Y. Baldwin, G.H.
Gear, & L. J. Lucito (Eds.), Educational planning for the gifted
Overcoming cultural, geographic, and socioeconomic barriers. Reston
VA:Council for Exceptional Children, 1978.
Renzulli, J. S., & Hartman, R. K. Scale for rating behavior characteristics
of superior students. Exceptional Children, 1971, 38, 243-8.
Renzulli, J. S., Reis, S., & Smith, L. The revolving door identification
model. Wethersfield, CT: Creative Learning Press, 1980.
Roedell, W. C., Jackson, N. E., & Robinson, H. B. Gifted young children.
NYC: Teachers College Press (Columbia University), 1980.
Stanley, J. C. The predictive value of the SAT for brilliant seventh- and
eighth-graders. College Board Review, 1977-78, 106, 30-37.
Stanley, J. C., & Benbow, C. P. Using the SAT to find intellectually
talented seventh graders. College Board Review, 1981-82, 122, 3-7.
Stanley, J. C., George, W. C., & Solano, C. H. (Eds.). The gifted and
the creative: A fifty-year perspective. Baltimore: The Johns Hopkins
University Press, 1977.
Stanley, J. C., Keating, D. P., & Fox, L. H. (Eds.). Mathematical talent
Discovery, description and development. Baltimore: The Johns Hopkins
University Press, 1974.
Terman, L. M. Mental and physical traits of a thousand gifted children
Genetic studies of genius, (vol. 1). Stanford: Stanford University Press
VanTassel-Baska, J. The Illinois state-wide replication of the Johns
Hopkins study of mathematically precocious youth. In C. P. Benbow
& J. C. Stanley, Academic precocity: Aspects of its development.
Baltimore: The Johns Hopkins University Press, 1983.
VanTassel-Baska, J., & Prentice, M. The midwest talent search: Catalyst
for state-wide efforts on behalf of the precocious. Roeper Review.
Contributed by: Parent on 10/23/2006
I feel that the diagnositic aspect of the talent search should be more widely considered by schools. I wonder if equally detailed information can be generated from the SCAT above level test that younger students take.
Contributed by: Parent on 2/20/2005
This is a very good paper, however, as the other comment here mentioned, and I agree, there should be an updated version to add to this work. The progress or lack thereof should be noted in an ammended version.