Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13

Multiple Intelligences: From the Ivory Tower to the Dusty Classroom-But Why?

by Mindy Kornhaber - 2004

This article draws on research conducted over a 10-year period in an attempt to answer three central questions about the widespread adoption of Gardner's theory of multiple intelligences (MI): Why do educators adopt MI? Once MI is adopted, does anything really change in practice? When educators claim MI is working, what is happening in practice?

Howard Gardner’s (1983) theory of multiple intelligences (MI) has now been adopted and implemented for use in schools on six continents, from grade levels spanning pre-kindergarten through college, and for an enormous diversity of student populations: ‘‘typical,’’ special needs, gifted, juvenile delinquents, and adult learners.

That educators have so readily taken to MI is quite remarkable. The book in which Gardner first published the theory, Frames of Mind, offers perhaps six paragraphs of direct information on applications (see Gardner, 1983, pp. 386–392, passim). The index lacks entries for such educational mainstays as curriculum, instruction, pedagogy, teachers, or testing. Furthermore unlike other major reform efforts-such as Henry Levin’s Accelerated Schools (1999), Theodore Sizer’s Coalition of Essential Schools (Muncey & McQuillan, 1993), Robert Slavin’s Success for All (Slavin, Madden, Dolan, & Wasik, 1996), MI’s founder never established a field staff to support the implementation of his ideas or evaluate their use.

Given the absence of information and structure to support the theory’s implementation, it is reasonable to ask why do educators adopt MI? Furthermore, once MI is adopted, does anything really change in practice? And, when educators claim MI is working, what is actually happening in practice? This article seeks to address these three questions, drawing on research conducted over a period of about 10 years.


Two investigations help shed light on the question of why educators adopt MI (Kornhaber, 1994; Kornhaber & Krechevsky, 1995). Kornhaber and Krechevsky (1995) provided an initial description of the state of MI in educational practice. The study’s sample consisted of 9 diverse sites. The sample was in not random. Instead, it was drawn from a collection of site- initiated correspondence with Howard Gardner or his associates at Project Zero or from information reported to Gardner by visitors to the sites. To be selected each site was required to have been using MI in classrooms for at least 2 years. The sites ranged in scope from a single teacher who had been intensively using MI since shortly after the theory’s publication to several schools within a large, northeastern urban school district in which central office staff had supported the theory’s adoption. One site was a middle school. The rest served elementary-age populations.

Data came from interviews with the primary implementers at each site (usually several teachers and the principal; on occasion a central office staff member), documents from the sites, and observations of classrooms. The interview consisted of questions about the site’s student population and surrounding school district, about how the educator first encountered and implemented MI, and about approaches to curriculum, assessment, pedagogy, materials, parent conferences, school structure, and organization. Ninety-one educators were interviewed either in person or by phone.

Interviews were recorded and transcribed, and handwritten observations and observer notes were typed into word processing programs to facilitate coding and data analysis. Data analysis was continuous, beginning with conversations at the start of data collection between the researchers that were aimed at describing, comparing, and contrasting what each had heard and observed. This continued through initial readings of field notes and interview transcripts, through the generation of an indexing system for each transcript or note, and into the condensing of these indices to arrive at a cluster of themes and subthemes. Data were then coded to reflect these themes. These themes, in turn, enabled an initial, synthetic description of what had transpired across the different sites.

This study revealed that educators in the sites adopted the theory for a well-defined set of reasons:

MI validated what educators already know. Educators’ everyday observations were already aligned with the theory’s idea that people learn in a variety of ways. That a Harvard psychologist’s theory validated their experiences encouraged educators to learn more about the theory and consider the theory’s implications for practice.

MI complemented educators’ existing philosophies and beliefs. For example, MI aligned well with constructivist and progressive philosophies that children learn through activity, and with beliefs that it is important to educate the whole child, that all children have gifts, and that all children should experience success in at least one area.

Educators already used some practices that fit with the theory. These included project-based curriculum, arts-integrated approaches, thematic units, learning centers, and hands-on learning, each of which tends to draw on a range of intelligences.

MI provided a framework for organizing educators’ practice. Teachers, like skilled practitioners in other disciplines, have a large repertoire of knowledge and methods that they draw on intuitively (Polanyi, 1958; Schön, 1983). MI offered teachers a handy way of categorizing and understanding the contents of their repertoire. In essence, it has a closet- organizing effect (Kornhaber, 1994, 1999; Kornhaber & Krechevsky, 1995).

Educators reported that MI helped extend their practice. When a closet organizer is put in place, it is possible to see more clearly both what has been categorized and also what is missing. As one teacher put it, MI ‘‘allows us to work from our heart and our head together.’’ In effect, MI provided educators with a framework that supported systematic reflection on, and development of, their practice.


While the previous investigation helped to clarify why MI is adopted, it is not necessarily the case that adoption contributes to any change in practice. To explore whether MI was associated with change, a small study that drew on a subsample of three sites from the first investigation was conducted

(Kornhaber, 1994). The aim of this study was to test, on a limited basis, the following null hypothesis: MI makes no difference in practice. Given that teachers already had practices and beliefs that aligned with the theory, and that MI does not stipulate the adoption of any particular practices, it was possible that MI was a name-only phenomenon. Teachers might simply say they ‘‘did MI,’’ while actually only doing whatever it was they had always done.

The three sites in the sample were selected because they provided a ‘‘best case scenario’’ (Light, Singer, & Willett, 1990, p. 225). The sample contained two public elementary schools and one public middle school that served primarily middle class, white students. Each had highly experienced teachers, good support for the arts, and a curricular focus that went beyond basic skills. If MI were going to have some impact, it might be most readily found in schools where issues of funding, staffing, poverty, and community support were not problematic.

To carry out this study, transcripts of interviews from teachers and the principal for each school were coded in a trivalent way, as either consistent with prior practice, an extension of prior practice, or a change from prior practice in four areas: curriculum, assessment, pedagogy, and school structure. Coded text for each of these areas was placed in separate matrices. For each of the four areas in each school, this yielded a visual display of stasis, extension, or change.

This analysis revealed that for each of the three schools there was at least extension or change in two or more of the four areas. Further, the study suggests that the extent to which change occurred was related to the kind of practice that existed in the school prior to the adoption of MI. To illustrate, School 1 had a very experienced staff that had jointly studied constructivism for several years and that shared a philosophy of educating the whole child. In addition, the school provided Suzuki and regular music classes and had an arts-enriched, thematic curriculum. MI fell well with this school’s zone of proximal development (Vygotsky, 1978), and in it there was more an extension of existing practice, rather than outright change. In contrast, School 3 was initially a fairly traditional elementary school, in which textbooks and teacher-centered instruction were the mainstays. A group of teachers formed a school-within-a-school to focus on MI. For that group, MI helped to foster marked changes in curriculum, assessment, and pedagogy. These changes eventually migrated throughout the school. In sum, this small study (as well as many informal reports from teachers) indicated that MI is not just a faddish label for existing practice but instead spurs educators to develop their practice.


There are many testimonials by educators that MI is helpful to students. However, such claims have several attendant mysteries. One is that, in the absence of a prescription for implementing MI, it is unclear what practices educators are employing to make the theory useful. Yet another is that it is unclear what benefits educators have in mind and, more particularly, how prevalent any particular benefit may be. The Project on Schools Using MI Theory (SUMIT) was a 3.5-year study developed to explore such issues (Kornhaber, Fierros, & Veenema, 2004).

Data collection for SUMIT took place in two phases: The first phase consisted of telephone interviews with the principals (or other school leaders) of 41 diverse schools. Thirty-nine were public schools. One-third of the schools had between 40–100% of the students receiving free and reduced meals. One quarter of the schools had student populations that were between 50–100% African American or Latino, or a combination of both. The schools were located in urban, suburban, and rural areas across 18 states and one Canadian province.

This sample was culled from an initial population of over 60 schools that had either contacted Gardner with reports that MI was beneficial or that had been referred by consultants who felt that MI was working well in those schools. Forty-four schools met our initial criterion of having used MI in classrooms for at least 3 years. Forty-one both agreed to participate and completed the phone interview.

The second phase of data collection entailed visits to 10 schools to observe classroom practice, interview teachers, and document examples of student work. All 10 schools were regular (i.e., noncharter, nonmagnet) public schools, and most were elementary. These schools received one or more external awards for excellence and were chosen to reflect, en masse, a great diversity in student populations.

The telephone interview addressed four areas: the school’s students, staff, and community; the methods used to introduce the school to MI; curriculum, assessment, and organizational practices, and outcomes associated with the use of the theory. Because MI is typically not the only idea influencing a school (Kornhaber & Krechevsky, 1995), when interviewees reported outcomes, we probed their responses to tease out whether the change was one the respondent felt was associated with the school’s adoption of MI or whether it was unrelated to MI. The interviews were tape-recorded and transcribed verbatim.

Data analysis was continuous, beginning with discussions that reviewed each interview and continuing through the formulation of themes and subthemes, to the evolution of a coding scheme. Interrater reliability of the coding scheme for two raters was established at .83 (Cohen’s Kappa). Then coding of data began by tackling the outcomes reported during the 41 interviews.

Coded transcripts revealed that the schools reported positive associations between MI and four outcomes:

Nearly 80% of the schools reported improvements in standardized test scores, of which nearly half of the schools associated the improve- ment with MI.

80% reported improvements in student behavior, with slightly more than half associating this improvement with MI

80% reported increased parent participation, with 60% associating the increase with the school’s adoption of MI.

80% reported a range of improvements for students with learning disabilities (e.g., improved learning, improved motivation, effort or social adjustment), with all but one of the schools associating this improvement with MI.

This self-report data can be bolstered by other lines of evidence (Neisser, 1986; Schwarz, 1999). There are several logical reasons that MI might be associated with such outcomes. For example, improvements in test scores might be spurred by uncovering the curriculum. Instead of plowing through textbooks, educators in many schools were striving to engage students through in-depth units that employed a range of media and symbol systems. For instance, one teacher associated increased achievement with MI in this fashion: ‘‘I think children have more opportunities to achieve . . . because they have different modalities and different ways to express themselves.’’ That only about half the schools reported such gains is not surprising. Few of the schools sought to raise test scores through direct test preparation, but instead focused on what they considered best practice.

Improvements in student discipline can be logically linked to greater engagement among students whose strengths are outside the linguistic and logical-mathematical areas typically valued by schools. Educators also commented that the theory helped to support school cultures in which many different kinds of learners were valued. If students are more engaged academically and socially, then it makes sense that fewer students will get into trouble.

Parent participation is a notable outcome because it is associated with student achievement (Epstein & Connors, 1992; Henderson, 1987, 1994). One reason that MI may foster increased parent involvement is that the theory values a diversity of adult end-states, not only those requiring academic credentials. In schools that use MI, this may help widen the range of parents who feel comfortable participating in school activities. In the schools that participated in SUMIT’s study, parents from a wide variety of occupations were active in the school. For example, in a school in which the thematic unit was community, parents who were train conductors, police officers, store owners, and researchers came in with their typical uniforms and equipment to discuss their work. As one principal remarked, ‘‘MI has given us real reasons to have parent volunteers.’’

MI may be associated with benefits for students with learning disabilities because the theory supports the idea that these students had strengths and not only weaknesses. Acknowledging these strengths seemed to offer academic as well as emotional benefits. For instance, a principal remarked that students with learning differences ‘‘feel good about being able to choose and play on strengths, while they’re also working on weaknesses in other areas so that they can become effective.’’ Our classroom observations revealed that youngsters with special needs were commonly working constructively within regular classrooms and typically with the same high level of engagement as other students. As discussed later, the theory helped to build or support school cultures that were respectful and caring.

Once these benefits and their prevalence were identified, SUMIT’s research focused on surfacing the practices that the schools were actually employing. Our analysis of coded interview data and school observations reveal a set of six organizational practices that are commonly used across this very diverse set of schools. We call these Compass Point Practices and regard them as guidelines for orienting practice and charting progress to incorporate MI in a constructive fashion. The Compass Point Practices are as follows:

Culture. The cultures of the schools in our study shared four salient qualities that support diverse learners: First were deeply held beliefs that all children have strengths and can learn. Second, all members of the school community were to be cared for and respected. Third, the adults in these schools worked hard. Teachers invested a good deal of energy to develop new curriculum, share ideas, and create unusually rich classroom environments. Fourth, there was a sense of joy and great excitement about learning. In our 10 site visits, it was wholly typical to see classrooms of students who were eagerly and actively engaged.

Readiness. While many reforms advocate the sorts of cultures we found (e.g., Comer, 1988; Levin, 1999; Meier, 1985; Sizer, 1984), they differ with respect to the optimal speed of implementation. In the schools we investigated, MI typically moved into classroom practice gradually. There was a readiness period of 12 to 18 months between the introduction of the theory into the school’s conversation and its application in several classrooms. The theory gradually percolated into actual practice rather than being summarily added on top.

Collaboration. Teachers in these schools often had formal arrangements that facilitated collaborations, such as team teaching or common planning time. They also participated in abundant, informal, pick-up collaborations. For example, in a unit on Pittsburgh’s rivers, a classroom teacher spontaneously asked the art teacher if she might provide lessons to enable the students to sketch, as well as write up, their observations in the field. The result was much richer and more faceted student work. We believe that MI helped teachers to appreciate and draw on each other’s strengths. This, in turn, enhanced students’ learning experiences.


Choice. Teachers often allowed students to choose how they wanted to acquire and demonstrate knowledge and skills. However, students’ choices were also constrained: The choices had to be worthwhile to the teacher, and not only the student. A principal described this clearly: ‘‘We tell students, ‘The last four times you did . . . a visual representation. Let’s see if we can’t have you do something in another format.’’ Through controlled choice, a wide range of learners could be constructively engaged through their strengths and stretched to develop new know- ledge, skills, and forms of representation.

Tool. SUMIT and earlier research (Kornhaber & Krechevsky, 1995) reveal that when educators first start implementing MI, many of them attempt to teach everything in seven or eight ways. However, after 2 years, educators usually rethink this approach. As one principal put it, ‘‘We are way over that kind of thinking.’’ After this rethinking, educators started to draw more strategically on the theory to open up the curriculum to diverse learners. For example, they might provide opportunities sometime in the day or week for students to use musical intelligence, rather than making singing part of every single unit. In essence, they redirected the theory to support students’ learning, rather than use the school to support the theory. In only a very few schools was MI described or taught to students.

Arts. The arts played a vital role in nearly every school that was interviewed and in all the schools that were visited. The arts in these schools are essential as ends in and of themselves, and they also serve as a powerful means of learning and representing other disciplinary content. Most of the schools had at least part-time art and music teachers who taught the arts as separate disciplines. In addition, the regular classroom teachers also widely integrated the arts into their lessons, as the above- mentioned study of Pittsburgh’s rivers helps to illustrate.

The work carried out by SUMIT provides a much clearer idea of the practices used in schools that associate MI with benefits for students. Some of these practices, especially the creation of a caring school culture and the presence of teacher collaborations, are common to school reforms and to good school practice generally (Comer, 1988; Levin, 1999; Meier, 1995; Sizer, 1984). At the same time, the Compass Point Practices have some notable points of departure. One of these is the theory-based framework for acknowledging and working with a range of intellectual strengths. Another is the prominent and vital role for the arts.

The identification of the Compass Point Practices is no more likely to shift practice in schools than is any other research-based framework. Even if this framework entered educational practice on a broad scale, there is little reason to suspect that it will broadly improve schooling. The case of MI itself illustrates the problem of widespread adoption, with equally widespread quality of practice. The few paragraphs that Gardner (1983) provided in Frames of Mind describing how the theory might be applied were inadequate to guide practice. Therefore, to support educators’ efforts to employ the Compass Points and MI as wisely as possible, SUMIT has also produced a set of detailed examples illustrating how these frameworks work in schools and classroom practice (Kornhaber, Fierros & Veenema, 2004). These examples clearly describe teacher strategies, materials, room arrangements, curriculum activities, and assessment practices. It is hoped that these examples, alongside theory and research, will be useful to educators in their efforts to use MI to advance diverse learners’ understanding.


Comer, J. (1988). Educating poor minority children. Scientific American, 259(11), 42–48.

Epstein, J. L., & Connors, L. J. (1992). School and family partnerships. The Practitioner, 18(4), 3–10.

Gardner, H. (1983). Frames of mind: The theory of multiple intelligences. New York: Basic Books.

Henderson, A. (1987). The evidence continues to grow: Parent involvement improves student achievement: An annotated bibliography (National Committee for Citizens in Education Special Report). Columbia, MD: National Committee for Citizens in Education.

Kornhaber, M. L. (1994). The theory of multiple intelligences: Why and how schools use it. Unpublished manuscript, Harvard Graduate School of Education, Cambridge, MA.

Kornhaber, M. L. (1999). Multiple intelligences theory in practice. In J. H. Block, S. T. Everson, & T. R. Guskey (Eds.), Comprehensive school reform: A program perspective, (pp. 179–191). Dubuque, IA: Kendall/Hunt.

Kornhaber, M. L., Fierros, E., & Veenema, S. (2004). Multiple intelligences: Best ideas from theory and practice. Needham Heights, MA: Allyn & Bacon.

Kornhaber, M. L., & Krechevsky, M. (1995). Expanding definitions of teaching and learning: Notes from the MI underground. In P. Cookson & B. Schneider (Eds.), Transforming schools. New York: Garland Press.

Levin, H. (1999). Learning from accelerated schools. In J. H. Block, S. T. Everson, & T. R. Guskey (Eds.), Comprehensive school reform: A program perspective (pp. 17–32). Dubuque, IA: Kendall/Hunt.

Light, R. J., Singer, J. D., & Willett, J. B. (1990). By design: Planning research on higher education. Cambridge, MA: Harvard University Press.

Meier, D. (1995). The power of their ideas: Lessons for America from a small school in Harlem. Boston: Beacon Press.

Neisser, U. (1986). Nested structure in autobiographical memory. In D. C. Rubin (Ed.), Autobiographical memory (pp. 71–88). Cambridge, UK: Cambridge University Press.

Polanyi, M. (1958). Personal knowledge: Towards a post-critical philosophy. Chicago: University of Chicago Press.

Schön, D. (1983). The reflective practitioner: How professionals think in action. New York: Basic

Schwarz, N. (1999). Self reports of behaviors and opinions. In N. Schwarz, D. Park, B. Knauper, & S. Sudman (Eds.), Cognition, aging, and self reports (pp. 17–43). Philadelphia: Psychology Press.

Sizer, T. (1984). Horace’s compromise: The dilemma of the American high school. Boston: Houghton- Mifflin.

Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard University Press.

Cite This Article as: Teachers College Record Volume 106 Number 1, 2004, p. 67-76
https://www.tcrecord.org ID Number: 11510, Date Accessed: 12/2/2021 1:40:08 PM

Purchase Reprint Rights for this article or review
Article Tools
Related Articles

Related Discussion
Post a Comment | Read All

About the Author
  • Mindy Kornhaber
    The Pennsylvania State University
    E-mail Author
    MINDY L. KORNHABER is an assistant professor in the Department of Education Policy Studies at the Pennsylvania State University’s College of Education. Her research explores how human potential can be developed both to a high level and on an equitable basis. Her most recent publications include ‘‘Assessment, Standards and Equity,’’ a chapter for the new edition of The Handbook of Multicultural Education, edited by James A. Banks and Cheryl A. McGee Banks, and a book, coauthored with Edward Fierros and Shirley Veenema, titled Multiple Intelligences: Best Ideas from Research and Practice (Allyn & Bacon, 2004).
Member Center
In Print
This Month's Issue