Abstract
There is substantial evidence on the effectiveness of many forms of advanced education, especially various approaches to acceleration, ability grouping, and curricular innovations such as structured curriculum and enrichment. Nonetheless, additional research on the ways in which advanced education impacts the learning and lives of students across the variables of class, race, ethnicity, and gender is still needed, as it is for most educational interventions. Jonathan Plucker and Carolyn Callahan share the evidence base for several popular strategies and describe what evidence is still needed.
Keywords
Research shows that talent development and gifted education are effective, but more research is needed into how well they work for specific populations of students.
For decades, Americans have held conflicting views about the wisdom of offering advanced educational courses of study — such as honors classes, Advanced Placement (AP) courses, and special programs for “gifted” students — in the public schools. On one hand, supporters argue that these programs benefit not just the students but all of us, since they identify and nurture talented individuals who will go on to play leading roles in society. On the other hand, critics decry advanced programming as elitist, inequitable, and (on balance) harmful to the country, since they tend to benefit children who are already advantaged, while taking resources away from those who can’t participate.
In recent years, policy makers and the media have generally taken a positive view of advanced education programs, especially AP, which has seen its enrollments grow rapidly across much of the country, more than doubling between 2009 and 2019. Today, however, advanced programs are coming under fire once again, especially in urban school districts, and with particular urgency given the present economic crisis and heightened concerns about systemic racism.
Much of the current criticism is familiar from past debates: “Those kids will be fine on their own, without special programs.” “You shouldn’t separate students based on ability; teachers can just differentiate for every student in the regular classroom.” “Advanced programs are biased against lower-income students.”
The counterarguments, too, are familiar: “All children should have the opportunity to learn something new in school every day.” “Curriculum should be challenging to all students.” “Students who come to school learning well above grade level should be given the opportunity to go beyond.”
From our perspective, advocates make the stronger case.
But a new line of criticism has emerged, as well: “Research says that advanced programs don’t work.” Often presented as a known fact (e.g., Keep, 2019), this assertion has become increasingly common, especially among those who’ve called for the elimination of exam-based school admissions and other selective programming, most notably in New York City and Seattle, Washington. For example, in an influential August 2019 report, the NYC Student Diversity Advisory Group (SDAG, 2019) repeatedly claimed that there is no evidence that gifted programs are effective; instead, the report argues, the New York City schools should offer “enrichment programs” (also see Potter & Burris, 2019). However, given that enrichment is a form of gifted education, this is akin to arguing that pain medicine doesn’t work, and that people should use ibuprofen instead.
These attempts to dismiss the benefits of advanced programs tend to be thinly sourced, if they contain any supporting research at all. For example, the SDAG report has 37 citations, the majority of which are newspaper stories, think pieces, or a report of the SDAG — in other words, not research. By contrast, where analysts have taken a careful and even-handed approach to reviewing the evidence in this area (as in, for example, the recent report commissioned by the Massachusetts Department of Education; Ansel, 2019), they have arrived at generally positive conclusions about the effectiveness of advanced education.
Consideration of what kinds of programs we should or should not offer in public schools — by school boards, superintendents, state legislatures, or the general public — may well lead to the conclusion that the pros of advanced programming are outweighed by the cons, or that other investments should have priority. However, if the debate is to be rational, it is important that those who weigh in are informed about the existing defensible evidence rather than relying on false, one-sided assertions or biased “popular” claims.
Reviewing the evidence
Despite the recent wave of criticism, substantial research evidence supports the conclusion that advanced learning programs are effective. Does high-quality, gold-standard, replicated research exist for every possible intervention that falls under the umbrella of gifted education or talent development? Of course not. For many reasons, it is an anomaly when an educational program or practice has gold-standard evidence based on comparing student outcomes between groups who have been randomly assigned to receive or not receive that educational intervention. Consider, for example, that despite spending billions to study how best to help children learn to read — one of the most foundational tasks in all of education — we continue to debate the issue (see Goldstein, 2020).
However, if we lack gold-standard research, that doesn’t mean we lack evidence. In fact, there is a great deal of intervention research regarding the effectiveness of programs for advanced learners. Below, we summarize the evidence base for various types of interventions. Because randomized experimental studies aren’t available in large numbers for every intervention, we rely on studies offering a preponderance of evidence that an intervention is effective, with a bias toward experimental studies when available.
Also, we have chosen to focus on studies that measure the effectiveness of specific educational models and interventions. Some researchers have attempted to examine “gifted programs” in general, and they have found that gifted education produces meager effects (e.g., Adelson, McCoach, & Gavin, 2012; Bui, Craig, & Imberman, 2014). However, that makes no more sense than concluding that “science education” has meager effects. One has to ask: What kind of science education? How was it being taught? What curricula are we talking about? How are the teachers prepared? Asking these kinds of specific questions about specific programs — whether we’re discussing science education or gifted education — is much more useful than lumping together a wide variety of interventions and trying to measure their overall effectiveness.
The following synopses provide overviews of the available research support for specific kinds of advanced educational programs, roughly in order from most to least support. Although we assume that the interventions studied below were well-
designed and carefully implemented, we caution that not every strategy works the same every time, even in similar contexts, for every student.
Strategies that have the strongest research base
Acceleration
One of the most-studied intervention strategies for advanced learners, acceleration refers to moving particular students forward more quickly than is typical, such as by having them enter kindergarten early, skip grades, study a telescoped curriculum (i.e., compressing material into a shorter time frame), or participate in dual enrollment or early college programs. Researchers have gathered overwhelming evidence that acceleration has positive effects on student achievement, and the evidence also dispels the myth that acceleration is detrimental to students’ social engagement or emotional well-being. No doubt, the research base could be strengthened by additional experimental studies, as well as evaluations of acceleration’s effects on diverse populations. Overall, though, evidence that acceleration increases student learning is hard to question, at this point.
The list of compelling studies is far too long for us to include here, but for readers who want a more complete summary of the research findings, we recommend the Nation Empowered report edited by Susan Assouline and colleagues (2015) and the meta-analysis by Saiying Steenbergen-Hu, Matthew Makel, and Paula Olszewski-Kubilius (2016).
Ability grouping
Although the evidence of the effectiveness of ability grouping is not as clear cut as the data supporting acceleration, researchers have provided convincing evidence that flexible ability grouping (i.e., grouping that can change as student abilities and needs change) is a net positive for the learning of our most and least advanced students. Of course, the important distinction here is the contrast between the concept of flexible grouping and tracking. Flexible grouping (often implemented as cluster grouping; see Gentry, 2018) provides students with opportunities to be grouped and regrouped for instruction across and between different academic areas, with student placements shifting across time as developmental differences emerge or a student responds differentially to instruction. Tracking, on the other hand, involves placing students in rigid groups from which they rarely move, a practice for which there is little research support.
In a comprehensive analysis of existing research on this topic, Steenbergen-Hu and colleagues (2016) found that within-class ability grouping was the most effective grouping strategy for promoting advanced learning, and the research suggests students in the top and bottom achievement quartiles may benefit the most from ability grouping, with students in the middle quartiles experiencing little to no benefit or deficit (Plucker & Callahan, 2014). For the positive effects to appear, students must receive an appropriately differentiated curriculum, by, for example, being assigned tasks such as reading poems that are a higher level of complexity or require deeper levels of understanding or simply learning activities from a more advanced grade level.
Curricular interventions
A growing body of evidence suggests implementation of pre-differentiated, prescriptive curricula for advanced learners — those who have provided evidence of excelling in particular domains of learning — tends to result in more advanced outcomes than grade-level curricula or less-structured enrichment opportunities (Plucker & Callahan, 2014). These pre-differentiated curricula are characterized by clearly stated learning outcomes, strategies for formative assessment accompanied by direction on how to use data effectively in modifying curricular and instructional decisions, lesson plans that reflect the modification based on student data, and assessments that reflect the varied learning levels. Both standardized achievement tests and specific curricular assessments have shown that students who were randomly chosen to participate in lessons using these curricula achieved at higher levels than did similar students who were offered standard lessons. To date, strong evidence for these benefits has been found in a number of subject areas — science, social studies, reading, language arts, and mathematics (e.g., Callahan et al., 2015; Gavin et al. 2009; VanTassel-Baska, 2002; VanTassel-Baska, 2009) — as well as in homogeneous classrooms, pull-out classes, and heterogeneous classrooms. Several of the curricula also show promise in boosting the achievement of advanced students in underserved settings, specifically those in Title I classrooms (VanTassel-Baska et al., 2009) and in low-income, rural schools (Callahan et al., in press).
Strategies that have strong evidence, but require additional research
Enrichment
Enrichment programs with a focus on the development of process skills (creative thinking, higher-level thinking skills, etc.) and with a focus on language arts at the elementary level are among the most widely used forms of advanced education (Callahan, Moon, & Oh, 2017), but the research literature includes relatively few third-party or experimental studies — although that situation has been improving (see Reis et al., 2011). The available research is mixed but generally promising, especially for residential summer enrichment programs such as those offered on university campuses or at satellite locations but operated by such programs as the Center for Talented Youth at Johns Hopkins University or Center for Talent Development at Northwestern University (e.g., Aljughaiman & Ayoub, 2013; Hany & Grosch, 2007; Kim, 2016). To date, though, there have been few high-quality research studies on the value of enrichment programs for underserved populations.
We suspect that it would be particularly helpful if researchers were to study enrichment programs using experimental methods to determine whether they have positive impacts on a range of highly desirable soft skills (such as communication, collaboration, creativity, and the development of personal interests) without any negative impact on student test scores. We suspect this will be the case, but given the mixed research on enrichment to date, this research needs to be prioritized. Given the interest-driven nature of enrichment programs, this is likely one of the best, if not the best, interventions for developing creativity, and if researchers were to find that it has no negative effects, that would make it an even more attractive strategy for advanced learning.
Advanced Placement and International Baccalaureate
Research on Advanced Placement and International Baccalaureate provides ample positive data from surveys of teachers and students (Callahan & Hertberg-Davis, 2018), but the experimental research is mixed in quality (often failing to control for demographic and aptitude variables) and has mixed results regarding student outcomes (e.g., Dickson, Perry, & Ledger, 2018). Although it appears that students who were successful in AP classes are more likely to be successful in college, the predictive strength varies by course. And there are signs that these courses may be detrimental for some students who do not achieve above minimal AP test scores.
It is commendable that many states have implemented policies designed to help more students of color to participate and succeed in AP classes, and the work of Chester Finn and Andrew Scanlan (2019) offers case-study evidence of positive effects of AP participation for poor and minority students. What we need next, however, are more experimental studies that can provide solid evidence of the effectiveness of these policies and associated practices and that can help determine what will be most beneficial for the greatest number and range of students.
A strategy supported by few causal studies
Selective high schools
Among the oldest strategies for advanced education are public high schools that selectively choose high-performing students based on entrance exam scores (hence the label “exam schools”). However, despite the long history of this approach to advanced learning, very few experimental studies have been conducted on such schools, and other sophisticated research designs have produced mixed-to-negative results regarding whether attendance at selective high schools improves student outcomes (Abdulkadiroğlu et al., 2014; Dobbie & Fryer, 2014). In essence, the research issue for selective schools is not whether their students have impressive accomplishments (e.g., admission to highly selective colleges, major academic awards, high SAT and ACT scores) but whether their students would not have the same accomplishments if they attended another, non-exam high school.
The issues surrounding exam schools are complex and controversial (Finn & Hockett, 2012), replete with long-standing questions about lack of diversity in these schools, their value-added for students, and their ability to close excellence gaps, but advocates and policy makers should not expect the research to provide much insight in this area.
Strategies with little evidence of effectiveness
Heterogeneously grouped differentiation
Given the strategies and evidence described above, educators and policy makers can reasonably conclude there are research-supported interventions to promote advanced learning. However, most students who would benefit from advanced education spend most of their time in heterogeneous classrooms, with some having access to a pull-out classroom for a very limited time period each day (typically less than an hour per week) with a focus on enrichment learning activities (Callahan, Moon, & Oh, 2017). The heterogeneous classroom approach relies heavily on teachers’ willingness, skill, and time to differentiate curriculum and instruction for the wide range of student performance levels in their classrooms, including students who want and need a greater challenge (Peters et al., 2017). However, there is evidence that only a limited cadre of teachers are effective at doing so (Hertberg-Davis, 2009), and few studies have produced evidence that differentiation has a positive effect on student learning (Ober, 2016). Many teachers themselves appear to recognize how difficult it is to provide this sort of instruction effectively (Loveless, Parkas, & Duffett, 2008).
For their part, researchers note that the evidence base is thin in this area. For example, little is known about how working in heterogeneous classrooms affects teachers’ sense of efficacy, whether teachers improve over time at providing differentiated instruction, and more (Subban, 2006). Indeed, in our experience, those who advocate for differentiation as the primary strategy for addressing academic diversity have little to offer as evidence of its effectiveness beyond lists of districts that are trying this approach. And, as with other strategies for providing advanced education, there is little to no research on the effects of instructional and curricular differentiation on outcomes for racially, ethnically, or socioeconomically diverse students.
Psychosocial interventions
Increasingly, in recent years, we’ve heard educators make the claim that advanced education programs are no longer necessary in their school district, given the benefits of newly adopted psychosocial interventions — such as those targeting growth mindset, grit, or learning styles. Moreover, we continue to hear educators make the familiar argument that advanced education can cause psychosocial harm to students (e.g., Boaler, 2018; Matthews & Foster, 2013; Scott, 2013).
However, a review of the relevant research literature concluded that psychosocial interventions such as those noted above are not suitable for routine use in schools and may not work at all to promote advanced achievement and close excellence gaps (Plucker & Peters, 2018). Since that review, the research has turned significantly more negative, with major studies and reviews finding mixed to little evidence that psychosocial interventions provide learning benefits for any group of students — and especially advanced students (e.g., Burgoyne, Hambrick, & Macnamara, 2020; Burnette et al., 2020; Gandhi & Raver, 2020; Sisk et al., 2018). Even though these interventions (as well as differentiation) are often proposed as alternatives for strategies with a more solid research base, they do not, on their own, offer the level of sustained challenge that enables advanced students to develop growth mindsets and progress toward reaching their potential.
A path forward
Do we need more and better research into advanced education programs, especially on how to ensure they work for students of different races, ethnicities, genders, and economic backgrounds? Absolutely, but that can be said about almost any educational intervention. For now, educators, policy makers, and advocates should be aware of and confident in the empirical support for some specific approaches to providing advanced education — and the lack of support for others.
In addition, educators and researchers should avoid pitting the various interventions against each other, as though one is superior to all others, even though, in practice, these strategies may overlap considerably. Instead of seeking the one best strategy for every situation, it may be better to ask which set of strategies will be best in a given context for a given student or group of students. As the research base continues to grow, we’ll be more and more able to answer these questions.