Research Evidence Regarding the Validity and Effects of Talent Search Educational Programs
A major contribution to the field of gifted education has been the educational programs that have been spawned as a result of the massive identification efforts of the talent searches. The universities that offer talent searches have developed programs for these students. Many other institutions of higher education have replicated these models or developed their own. The programs are varied in structural features but all use SAT or ACT scores as entrance criteria. Other defining features include a compression of content matter into shorter periods of time (e.g., four years of mathematics compressed into two years of study or one year of mathematics compressed into a three week summer program) and early study of advanced content material. Most of these programs are summer programs but others are academic year or Saturday programs McCarthy and Brody, this issue).

This paper reviews the evidence about the validity of using talent search scores as an indication of ability and rate of learning for gifted students. It also reviews research regarding the effect of participation in special educational programs designed for talent search students, particularly short-term effects on high school achievement and long-term effects on college study and achievement.
Validity of the Talent Search Protocol for Selection Into Educational Programs
The talent search educational programs usually use SAT and ACT scores that are comparable to the average scores of college bound, high school seniors as entrance criteria. Thus, programs select junior high-aged children whose reasoning abilities are advanced by four to five years. Entrance scores are adjusted for the particular demands of the course; math and verbal scores may be used, for example, for courses that are thought to require aptitude in both areas. Higher scores may be required for courses that are very advanced and/or very compressed. The available research evidence suggests that these practices are valid.

Olszewski-Kubilius, Kulieke, Willis and Krasney (1989) found that SAT cutoff scores used to select students into fast-paced summer literature classes (in which 120 hours of honors level, high school instruction was compressed into 75 hours) were appropriate as measured by performance on standardized achievement tests. In summer self-paced mathematics classes, achievement was also high and it was found that students who scored slightly below cutoff scores (below 500 on SAT-Math) but had well developed study skills were able to cover as much material as students who met or exceeded the cutoff.

Bartkovich and Mezynski (1981) found that students who score at 600 or above on SAT-Math completed two high school level mathematics classes in just 50 hours of in-class instruction on the average. Similarly, junior high-aged students whose average SAT-M scores were above 600 achieved at high levels in a special program in which four years of high school mathematics was compressed into two and a half years (Benbow, Perkins & Stanley, 1983).

There is ample evidence that younger students can master advanced content material. Lynch (1992) found that junior high-aged students who took high school science classes, such as biology, chemistry or physics, within a three week summer program obtained average scores on standardized tests that were above the 70th percentile compared to high school students who typically take these tests after having one full year of instruction. Similarly, Kolitch and Brody (1992) report that all but a few of the talent search students who took high school or college level mathematics classes several years earlier than is typical received grades of A or B for those classes and excelled on the Advanced Placement calculus examination .

Talent search students who accelerate their coursework do not experience ill effects. There is no evidence of burnout (Kolitch and Brody, 1992) as students retain their interest in mathematics and continue to take courses. Learning mathematics in an accelerated class does not result in superficial learning nor does it negatively affect subsequent learning within the subject area (Mills, Ablard, & Lynch, 1992; Kolitch & Brody, 1992); students in fast-paced summer classes succeeded in subsequent classes as determined by their own reports and those of their teachers. Fast-paced classes are not detrimental to long-term retention of the subject matter (Benbow, Perkins and Stanley, 1983) as is evidenced by performance on standardized achievement tests taken long after the class is completed. Accomplishing high school coursework through fast-paced classes also does not affect college placement. Talent search participants who accelerated in mathematics via special programs were placed at an appropriate and advanced level in mathematics in college (Kolitch and Brody, 1992) unless they requested a special placement.

In summary, the research evidence suggests that talent search scores can provide a valid indication of level of developed reasoning ability and learning rate within several domains that can be matched to educational programs adjusted for pacing and content. While the research base on these issues is more substantial in the mathematical area (see Benbow, 1992 for a review) than the verbal area, the findings challenge widely held ideas about the amount of instructional time that is needed for mastery of content material and placements in grades and courses based solely on chronological age.

Predictive Validity of Talent Search Scores

Talent search scores have high predictive validity also. Talent search participants who obtain scores comparable to college bound high school seniors generally continue a pattern of high achievement in high school and college, take more advanced and accelerated courses, earn more awards and honors, and have higher educational aspirations (Olszewski-Kubilius and Grant, 1996; Burton, 1998; Benbow & Stanley, 1983; Benbow & Arjmand, 1990; Barnett and Durden, 1993). Benbow (1992 ) demonstrated that individual differences on the SAT obtained in junior high were related to individual differences in accomplishments in high school and college. In a follow-up study of students who were identified as mathematically talented by virtue of their talent search scores at age 12, 52% of the males and 44% of the females were pursuing scientific/medical careers 10 years later (Benbow & Arjmand, 1990).

SAT scores have predictive validity within the gifted population as well. Among students in the top 1% of ability, those whose SAT scores place them within the top quarter achieved at a much higher level in high school, college and graduate school than students in the bottom quarter (Benbow, 1992). Benbow and Arjmand (1992) write, “This test (the SAT) can identify a pool of future scientists who might meet our nation’s technological needs.” (p. 59). And students who scored at high levels on the SAT in junior high continued their superiority on the SAT in high school obtaining average scores 200 points higher on SAT-M and 170 points higher on SAT-V compared to college bound seniors (Benbow, 1992).
Thus, the talent search can provide information that is not only an indication of current reasoning abilities and learning rate but a predictor of future accomplishments. Rarely has the field of education had such powerful predictive tools at its disposal.

Short- and Long-Term Effects of Talent Search Educational Programs
An important question about summer programs is their influence on students both short and long term. Proponents of talent search programs and special educational programs built on talent search scores assert many benefits to participation (Olszewski-Kubilius, 1987). Some of these have been documented empirically. The studies reported here were direct assessments of the effects of programs and involved comparisons between groups of participants and nonparticipants or between participants who took different courses or were in different kinds of programs.

Fox, Brody and Tobin (1985) and Brody and Fox (1980) assessed the impact of three different kinds of educational programs (an accelerative summer mathematics program, an in-school accelerated mathematics program and a career awareness program) taken in junior high on students’ coursetaking behavior and their attitudes in high school. Comparisons were made between programs and also to control groups of students with similar tested abilities. Girls in this study who participated in the accelerated mathematics summer program continued to be accelerated at grade nine compared to control boys and girls; however, that advantage was lost by grade 11. At grade 11, the summer program girls were on par with boys who had not been in a program but accelerated in mathematics compared to girls who had not been in a program. The authors conclude that the summer program helped talented females to keep up with talented boys who tend to accelerate without any intervention.
Barnett and Durden (1993) compared students who had participated in a talent search to students who had participated both in the talent search and in special summer programs. While both groups of students had a pattern of high achievement and completed a rigorous high school program, compared to the talent-search-only group, the students who participated in summer programs took more advanced courses and AP exams at an earlier age, were more likely to take the rigorous AP Calculus BC exam, took College Board Achievement Tests more frequently and earlier, and took more college classes while still in high school.

Similarly, Olszewski-Kubilius and Grant (1996) compared talent search participants who took mathematics in a summer program to students who took summer courses in other subjects. They found that females who studied math benefited more than students who took other subjects. The mathematics females tended to accelerate themselves more and earned more honors in math during high school than other summer boys or girls. An interesting finding of this study was that for females, participation in a summer mathematics program was associated with taking more AP courses in any subject. Thus, while the research on the effects of special programs suggests that participants generally pursue an accelerated and rigorous track within the subject of summer study, the effects may be generalized to other areas. This may be a result of increased confidence to succeed in rigorous academic settings.

There is also evidence that students who participate in a fast-paced mathematics class subsequent to a talent search participation are more likely to attend a more selective undergraduate institution (Swiatek and Benbow, 1991; Barnett and Durden, 1993) and to enter college earlier (Swiatek and Benbow, 1991). Females were also more likely to go on to graduate school (Swiatek and Benbow, 1991). Olszewski-Kubilius and Grant (1996) found that females who took a fast-paced mathematics class in the summer following talent search more often majored in math or science in college and had higher educational aspirations compared to students who studied other subjects.

In summary, participation in special educational programs subsequent to talent search can have many positive effects and these extend to high school and college coursetaking and educational aspirations. These effects, particularly potent for talented females, may be due to increased interest in the subject. However, it is more likely that achieving success in a class that is challenging, both because of the pacing and nature of the content matter and because of the capabilities of one’s classmates, does much to bolster confidence and raise expectations of oneself.

The effects of participation in talent search programs can also be less direct. Students who participate in talent search often are surprised at their performance on the SAT or ACT. They and their families become aware that their abilities in an area are exceptional. This may influence their choices of classes and extracurricular programs within school and result in a more rigorous educational program which can have profound benefits for students. Benbow and Arjmand (1990) differentiated a group of high and low academic achievers, based on college performance, within a group of students initially identified as mathematically talented through the talent search. They found that schooling variables, or the precollege curricula and experiences in mathematics and science prior to college, were the best predictors of differences between the two groups. Exposure to an academically rigorous educational program over a period of years is also associated with the development of abilities measured by the SAT and results in greater gains on SAT scores from junior high to high school (Brody & Benbow, 1990).

Thus, students who partake of special educational programs after participating in a talent search may reap the most benefits but the knowledge gained about one’s capabilities and how this can affect choices is also an important benefit.
There is ample strong research evidence to support the validity of the instructional models that have resulted from the talent searches. There is also evidence about positive impact of the talent search and talent search educational programs on students. Clearly this is one of the most successful accelerative models within the field of gifted education. Unfortunately the model is often perceived as appropriate only for a supplemental summer program. Talent search scores can be used effectively to select students for in-school, accelerated learning programs (see McCarthy; this issue). A continuing problem for talent searches and related education programs is access to them by economically disadvantaged students. Given the success of the model, educators needs to work to ensure that all qualified students have access to the testing program and supplementary educational programs.

References
Barnett, L. B. and Durden, W. G. (1993). Education patterns of academically talented youth. Gifted Child Quarterly, 37(4), 161-168.

Bartkovich, K. G. & Mezynski, K. (1981). Fast-paced precalculus mathematics for talented junior-high students: Two recent SMPY programs. Gifted Child Quarterly, 25(2, Spring), 73-80.

Benbow, C. P. (1992). Academic achievement in mathematics and science of students between ages 13 and 23: Are there differences among students in the top one percent of mathematical ability? Journal of Educational Psychology, 84, 51-61

Benbow, C. P (1992). Mathematical talent: Its nature and consequences. In N. Colangelo, S. G. Assouline, and D. L. Ambroson (Eds.), Talent development: Proceedings from the 1991 Henry B. and Jocelyn Wallace National Research Symposium on Talent Development (pp. 95 - 123). New York: Trillium Press.

Benbow, C. P. & Arjmand, O. (1990). Predictors of high academic achievement in mathematics and science by mathematically talented students: A longitudinal study. Journal of Educational Psychology, 82, 430-441.

Benbow, C. P., Perkins, S., & Stanley, J. C. (1983). Mathematics taught at a fast pace: A longitudinal evaluation of SMPY’s first class. In C. P. Benbow and J. C. Stanley (Eds.), Academic precocity: Aspects of its development (pp.51-78). Baltimore: Johns Hopkins University Press.

Benbow, C. P., & Stanley, J. C. (1983) Academic precocity: Aspects of its development. Baltimore: Johns Hopkins University Press.

Brody, L. E., & Benbow, C. P. (1990). Effects of high school coursework and time on SAT scores. Journal of Educational Psychology, 82, 866-875.

Brody, L. & Fox, L. H. (1980). An accelerative intervention program for mathematically gifted girls. In L. H. Fox, L. Brody and D. Tobin (Eds.) Women and the Mathematical Mystique. (pp. 164-178). Hillsdale, NJ: Erlbaum.

Burton, N. W. (1988). Young SAT-takers: Two surveys. Survey II: Test-taking history for 1980-81 young SAT-takers. College Board Report No. 88-1. New York, NY: College Entrance Examination Board.

Fox, L. H., Brody, L. and Tobin, D. (1985). The impact of early intervention programs upon coursetaking and attitudes in high school. In S. F. Chipman, L. R.

Brush and D. M. Wilson (Eds.) Women and mathematics: Balancing the equation. (pp. 249-274). Hillsdale, NJ: Erlbaum.

Kolitch, E. R., & Brody, L. (1992). Mathematics acceleration of highly talented students: An evaluation. Gifted Child Quarterly, 39, 78-86.

Lynch, S. J. (1992). Fast-paced high school science for the academically talented: A six-year perspective. Gifted Child Quarterly, 36(3), 147-154.

Mills, C. J., Ablard, K. E., & Lynch, S. J. (1992). Academically talented students’ preparation for advanced-level coursework after an individually-paced precalculus class. Journal for the Education of the Gifted, 16 (1), 3-17.

Olszewski-Kubilius P., and Grant, B. (1996). Academically talented women and mathematics: The role of special programs and support from others in acceleration, achievement and aspiration. In K. D. Noble and R. F. Subotnik (Eds.) Remarkable Women: Perspectives on Female Talent Development (pp. 281-294). Cresskill, NJ: Hampton Press.

Olszewski-Kubilius, P., Kulieke, M.J., Willis, G.B., and Krasney, N. (1989). An analysis of the validity of SAT entrance scores for accelerated classes. Journal for The Education of the Gifted, 13 (1), 37-54.

Swaitek, M. A., & Benbow, C. P. (1991). A ten-year longitudinal follow-up of participants in a fast-paced mathematics class. Journal for Research in Mathematics Education, 22, 138-159.