Bank Street's Developmental-Interaction Approach: A Century of Progressive Education in the Research Literature

The Bank Street approach, known formally as developmental-interaction, is among the oldest continuously practiced models of progressive early childhood education in the United States. Founded in New York City in 1916 as the Bureau of Educational Experiments, Bank Street built its pedagogy on a distinctive premise: that schools should be designed around the scientific study of children, and that cognitive growth and emotional growth are inseparable. This review traces the approach from Lucy Sprague Mitchell's experimental nursery school through Barbara Biber's theoretical synthesis, then examines the research record: the contested verdict of the federal Follow Through evaluation, Bank Street's own comparative studies, convergent evidence from research comparing child-initiated with didactic early education, and modern studies of Bank Street teacher preparation. The evidence supports the approach's core commitments — active learning, warm teacher-child relationships, and attention to the whole child — more firmly than it supports claims of superiority for the model as a package, a distinction this review keeps in view throughout.

In 1916, Lucy Sprague Mitchell (1878–1967), a Radcliffe-educated scholar who had served as the first dean of women at the University of California, Berkeley, founded the Bureau of Educational Experiments in New York City together with her husband, the economist Wesley Clair Mitchell, and the nurse-educator Harriet Johnson (Antler, 1987). The name was a manifesto. Deeply influenced by John Dewey's progressive philosophy, Mitchell believed that education should be built the way science is built: begin by studying how children actually grow and learn, design environments accordingly, observe what happens, and revise. In 1919 the Bureau opened a nursery school, led by Johnson, that functioned as a working laboratory in which teachers, psychologists, and researchers jointly studied young children and the environments that served them best (Bank Street College of Education, n.d.). Mitchell's Here and Now Story Book (1921) carried the philosophy into children's literature, arguing that the youngest children are best nourished by stories rooted in their own everyday world — kitchens, boats, streets, machines — rather than in fairy-tale kingdoms remote from their experience.

In 1930 the Bureau moved to 69 Bank Street in Greenwich Village, the address that eventually named the whole enterprise; in 1950, newly authorized to grant master's degrees, it became Bank Street College of Education (Bank Street College of Education, n.d.). The educational model itself acquired its formal name — the developmental-interaction approach — largely through the work of the psychologist Barbara Biber, who joined the Bureau in 1928 and led its research division for more than three decades, and whose career was devoted to viewing the child's school experience, in Zimiles's (1997) phrase, through a psychological lens.

Bank Street thus occupies an unusual position among early childhood traditions: it is simultaneously a school, a teacher-education institution, and a research organization, and it has been all three for over a century. That double identity is the reason there is a research record to review at all — and, as we shall see, the reason some of the most searching criticism of that record came from Bank Street's own researchers.

What "developmental-interaction" means

The canonical theoretical statement of the approach is Shapiro and Biber's (1972) paper in Teachers College Record, which drew together a half-century of practice into a coherent framework. The hyphenated name compresses two claims. "Developmental" signals that children's thinking changes qualitatively as they grow — the aim of education is not to accelerate children along a fixed track but to broaden and deepen the way they engage the world at each phase. "Interaction" carries a deliberate double meaning: it refers both to the child's active engagement with materials, ideas, and people, and to the interaction between cognitive and affective development — the conviction that thinking and feeling grow together and that school shapes both, whether it intends to or not (Shapiro & Biber, 1972; Nager & Shapiro, 1999).

The intellectual sources of this synthesis were eclectic. From Dewey came the view of education as experience reconstructed, of the classroom as a democratic community in miniature, and of curriculum as inquiry into the child's actual world — a lineage traced in careful detail by longtime Bank Street faculty member Harriet Cuffaro (1995). From cognitive-developmental psychology came stage-sensitive expectations about how children represent and reason. From psychodynamic theory came attention to identity, autonomy, impulse, and the emotional meaning of school life (Nager & Shapiro, 1999; Biber, 1984). What distinguished the resulting approach from a romantic child-centeredness was its insistence on the teacher as a trained observer: developmental knowledge plus disciplined observation of particular children, not sentiment, was to guide every classroom decision.

Biber's larger claim — elaborated across the forty years of writing collected in Early Education and Psychological Development (1984) — was that schooling has psychological consequences that reach far beyond measured achievement: it shapes children's sense of competence, their relation to authority, their willingness to risk error, and their disposition toward learning itself. That claim, controversial when the dominant metric was the achievement test, now reads as an early statement of what researchers today study under the headings of motivation, self-concept, and social-emotional learning.

The classroom: the here-and-now curriculum in practice

In practice, a developmental-interaction classroom is organized around what Mitchell called the "here and now": the child's immediate physical and social world, investigated firsthand and then represented in increasingly complex ways. Social studies — the study of how people live and work together — serves as the integrative core of the curriculum rather than one subject among many. Young children take trips into the surrounding neighborhood, observe the harbor or the bakery or the subway, and then reconstruct what they have understood in block buildings, paintings, dramatic play, writing, and discussion. Block building holds a place of special honor in the tradition; Cuffaro (1995) treats it as a paradigmatic Deweyan material, open-ended enough for children to embody and test their ideas about the world. Teachers keep detailed observational records, and assessment leans on documentation of children's actual work rather than on testing alone. The classroom is run as a community in which children practice negotiation, responsibility, and care — the democratic microcosm Dewey imagined.

The institutional embodiment of all this is the Bank Street School for Children in New York, a nursery-through-eighth-grade demonstration school that functions as a living laboratory for the college's graduate programs: student teachers observe and practice in classrooms that exemplify the approach, in a relationship between training school and laboratory school that has persisted since the Bureau's founding (Bank Street College of Education, n.d.). This arrangement — pedagogy, research, and teacher preparation under one roof — is itself one of Bank Street's most imitated ideas.

A national stage: Head Start, Follow Through, and a contested verdict

Bank Street's influence expanded dramatically in the 1960s. When the federal government launched Head Start in 1965, the college and its president, John Niemeyer, played an active role in the program's formation, ran training for Head Start staff, and established an early childhood model training center; the same year, the college published the Bank Street Readers, among the first mainstream basal reading series to portray a multiethnic, urban world to beginning readers (Bank Street College of Education, n.d.). From 1968 to 1981, Bank Street served as a sponsor in Project Follow Through, the federal "planned variation" experiment that extended Head Start into the elementary grades and set rival educational models to work in poor communities across the country.

Follow Through produced the single most cited — and most contested — piece of evidence about the Bank Street model. The national evaluation conducted by Abt Associates (Stebbins, St. Pierre, Proper, Anderson, & Cerva, 1977) compared children in each model's sites with comparison groups on a battery of instruments dominated by basic-skills achievement measures, across thousands of statistical comparisons. Its headline finding was unfavorable to the whole progressive family of models: programs emphasizing structured, direct instruction in basic skills (notably Direct Instruction and Behavior Analysis) outperformed the cognitively and affectively oriented models on the measured outcomes, while models like Bank Street's showed, on average, null or even negative effects on those measures relative to comparison classrooms. Honesty requires stating this plainly: on the evaluation's own terms, the developmental-interaction model did not demonstrate superiority at teaching disadvantaged children the mechanics of reading and arithmetic, and advocates of explicit instruction have cited that result ever since.

The evaluation itself, however, immediately came under heavy methodological fire. In a lengthy critique in the Harvard Educational Review, House, Glass, McLean, and Walker (1978) — who called Follow Through "the largest and most expensive federal educational experiment in this country's history" — argued that the outcome battery was narrowly weighted toward the goals of the basic-skills models, that models were poorly classified, that the covariance adjustments used to equate nonequivalent groups were unreliable, and, most strikingly, that variation in outcomes from site to site within a model generally exceeded variation between models: where and how a model was implemented mattered more than which model it was. On this reading, Follow Through mainly demonstrated that models were judged on tests aligned with some sponsors' goals and not others'.

Bank Street's own researchers had anticipated the measurement problem before the results appeared. Shapiro (1973), drawing on studies of young children conducted at Bank Street, argued that conventional achievement tests undersample the competencies that developmental programs aim to cultivate — initiative, reasoning in context, the ability to use knowledge flexibly — and that evaluating such programs requires observational and situational measures of competence, not test scores alone. One need not accept every element of the defense to see that both things can be true at once: the Follow Through data are genuine evidence that the model, as implemented in those sites, did not excel at conventional basic-skills outcomes, and the evaluation was genuinely ill-designed to detect the outcomes the model was built to produce. Fifty years later, that unresolved tension — what should early schooling be held accountable for? — remains the central question in evaluating progressive education.

What Bank Street's comparative research found

The most ambitious research to come out of Bank Street itself predates Follow Through. In The Psychological Impact of School Experience, Minuchin, Biber, Shapiro, and Zimiles (1969) studied nine-year-olds in four urban schools deliberately chosen to span the spectrum from traditional to modern educational philosophy, comparing the children's intellectual attainment and style, their perceptions of other people, their codes of social interchange, and their self-concepts, while also documenting, in effect, a natural history of four contrasting school cultures. Supported by the National Institute of Mental Health and running to more than five hundred pages, the study was a pioneering attempt to ask empirically what different kinds of schooling do to children beyond achievement — a question that school-climate researchers rediscovered decades later.

By modern standards the study's evidentiary force is limited: four schools, nonrandom assignment, and families who had self-selected their schools make causal inference impossible, and the same limitation applies to virtually all direct outcome research on Bank Street classrooms. What the comparative work established was less a verdict than an agenda — the demonstration that school environments differ measurably as psychological environments, and that children's self-understanding and social attitudes are legitimate outcomes of schooling. It is fair to say that direct, well-controlled evidence on Bank Street-model classrooms specifically remains scarce, and the honest reviewer must lean, as the next section does, on the broader research literature that tests the approach's core commitments.

Convergent evidence from the wider early childhood literature

Because pure model-versus-model trials are rare, the strongest available evidence bears on the developmental-interaction approach's central bets: that young children learn best through active, child-initiated engagement rather than didactic drill, and that early academic pressure carries motivational costs. Here the literature is more favorable. Marcon (1999), studying 721 four-year-olds across three cohorts in a large urban district, identified classrooms whose teachers held child-initiated, academically directed, or "combination" orientations — classified not by label but by cluster analysis of teachers' reported beliefs and practices — and compared children's development and learning across the three; children in child-initiated classrooms showed greater mastery of basic skills than children in academically focused programs, and the middle-of-the-road combination model fared worst on most measures. In a follow-up into the later elementary grades, children who had attended child-initiated preschools earned better grades by the end of elementary school, while the performance of children from academically directed preschools declined after fourth grade (Marcon, 2002). Stipek, Feiler, Daniels, and Milburn (1995), comparing 227 four- to six-year-olds in didactic versus child-centered programs, found that didactic programs produced higher scores on a letters-and-reading test but at a consistent motivational price: children in didactic programs rated their own abilities lower, expected less success, depended more on adults, showed less pride in accomplishment, and worried more about school.

The longest lens comes from the High/Scope Preschool Curriculum Comparison Study, in which 68 children living in poverty were randomly assigned to a scripted Direct Instruction preschool, a traditional nursery school, or the child-initiated High/Scope model. Academic and intellectual outcomes were essentially indistinguishable across groups for a decade — but by age 23 the Direct Instruction group showed markedly worse social outcomes, including substantially more felony arrests and treatment for emotional disturbance (Schweinhart & Weikart, 1997). The sample is small and the finding warrants caution, but its direction is consonant with the developmental-interaction claim that early schooling shapes more than skills. Meanwhile, the broad consensus review of preschool research assembled by Yoshikawa and colleagues (2013) concludes that high-quality preschool yields meaningful learning gains and that quality resides chiefly in stimulating, warm, responsive teacher-child interactions joined to intentional curriculum — a description that reads almost as a paraphrase of Bank Street doctrine, while also cautioning that unstructured, aimless programs achieve little. The same review documents what quality can deliver: well-implemented public preschool programs in cities such as Boston and Tulsa produced gains of between half a year and a full year of additional learning in early reading and mathematics (Yoshikawa et al., 2013). The fair synthesis: the evidence supports intentional, relationship-rich, child-active education over both drill and drift; it does not show that any one progressive brand outperforms another, and it leaves open the Follow Through-era worry that children who most need explicit instruction in code-based literacy skills should not be denied it.

The teacher as the unit of change: modern evidence

Because Bank Street's largest footprint is in teacher education, the most substantial recent research examines its graduates. A multi-year study by the Stanford Center for Opportunity Policy in Education — commissioned by Bank Street, a fact readers should weigh — surveyed graduates and employers, analyzed New York City administrative data, and conducted case studies of classrooms and schools (Darling-Hammond & Lit, 2015). Among its findings: 90 percent of surveyed employers rated Bank Street graduates well prepared; 87 percent of graduate respondents had remained in education; and graduates reported feeling significantly better prepared than comparison teachers for exactly the tasks the model prizes, such as building curriculum from students' experiences and interests (86 versus 54 percent) and developing students' questioning and discussion skills (83 versus 51 percent); graduates also reported greater confidence than comparison peers in their preparation to teach subject-matter content, by margins of roughly twenty percentage points in English language arts and mathematics. These are self-reports and employer perceptions rather than experimental effects on children's learning, and the commissioned design counsels modesty. Still, the study offers the clearest contemporary portrait of the developmental-interaction approach as a teacher-preparation philosophy: extended placements in classrooms, advisement built on the same observation-and-reflection discipline the model asks of teachers, and coursework anchored in child development (Darling-Hammond & Lit, 2015).

Legacy in the mainstream

Perhaps the strongest evidence of Bank Street's importance is how much of it has stopped looking distinctive. The ideas that early education should address the whole child; that young children learn through active engagement with materials, ideas, and people; that the teacher's first instrument is observation informed by developmental knowledge; and that classrooms should be emotionally supportive communities — all are now embedded in mainstream frameworks of developmentally appropriate practice and in the design DNA of Head Start (Nager & Shapiro, 2000; Bank Street College of Education, n.d.). The college's ventures beyond the classroom — the multiethnic Bank Street Readers, early educational software such as Bank Street Writer, and the science-adventure television series The Voyage of the Mimi — carried the here-and-now sensibility into publishing, computing, and broadcasting (Bank Street College of Education, n.d.). Nager and Shapiro's (2000) edited volume revisiting the approach argues that its continuing task is to hold developmental theory, cultural context, and equity together — an agenda that anticipated much of the current conversation in early childhood education.

Practical implications

For families and educators, several implications follow. First, the practices a visitor sees in a good developmental-interaction classroom — sustained block building, neighborhood study, teachers who write down what children do, curriculum that grows out of children's questions — are not decorative; they are the operational form of commitments that the broader research literature supports: active learning, warm and responsive relationships, and intentional curriculum (Yoshikawa et al., 2013; Stipek et al., 1995). Second, parents should distinguish between evidence for those ingredients, which is strong, and evidence that the Bank Street brand outperforms other coherent models, which does not exist; coherence and quality of implementation appear to matter more than label, a lesson Follow Through taught at national expense (House et al., 1978). Third, the motivational findings deserve real weight: didactic early programs can buy early letter knowledge at a measurable cost to children's confidence and independence (Stipek et al., 1995; Marcon, 2002). Fourth, the approach is not a license for drift — its founders insisted on disciplined observation and purposeful environments, and families should expect a school to articulate what teachers are watching for and how literacy and mathematics are systematically developed within the integrated curriculum. Finally, a program's willingness to say what the Follow Through record says — that progressive models have not proven superior on basic-skills tests, and that measurement of their distinctive goals remains hard — is itself a mark of the intellectual honesty the tradition claims as its birthright.

A century on, the Bureau of Educational Experiments looks like what it set out to be: an experiment that never stopped running. Its hardest test ended in a draw that each side still claims; its own researchers produced some of the most clear-eyed criticism of how such tests are scored; and its central ideas quietly won a broader victory by becoming the common sense of early childhood education. The developmental-interaction approach cannot yet point to the definitive trial its founders would surely have wanted to run. What it can point to is a convergent research literature affirming that thinking and feeling develop together, that children learn the world by acting on it, and that the teacher who watches closely teaches best — propositions first put to work in a Greenwich Village nursery school in 1919, and still holding up rather well under observation.

References

1. Antler, J. (1987). Lucy Sprague Mitchell: The making of a modern woman. New Haven, CT: Yale University Press. https://archive.org/details/lucyspraguemitch00antl
2. Bank Street College of Education. (n.d.). History of Bank Street. New York: Bank Street College of Education. https://www.bankstreet.edu/about-bank-street/history/
3. Biber, B. (1984). Early education and psychological development. New Haven, CT: Yale University Press. https://yalebooks.yale.edu/book/9780300039191/early-education-and-psychological-development/
4. Cuffaro, H. K. (1995). Experimenting with the world: John Dewey and the early childhood classroom. New York: Teachers College Press. ERIC ED386308. https://eric.ed.gov/?id=ED386308
5. Darling-Hammond, L., & Lit, I. (2015). The threads they follow: Bank Street teachers in a changing world. Stanford, CA: Stanford Center for Opportunity Policy in Education. https://educate.bankstreet.edu/books/12/
6. House, E. R., Glass, G. V., McLean, L. D., & Walker, D. F. (1978). No simple answer: Critique of the Follow Through evaluation. Harvard Educational Review, 48(2), 128–160. https://www.harvardeducationalreview.org/content/48/2/128
7. Marcon, R. A. (1999). Differential impact of preschool models on development and early learning of inner-city children: A three-cohort study. Developmental Psychology, 35(2), 358–375. https://doi.org/10.1037/0012-1649.35.2.358
8. Marcon, R. A. (2002). Moving up the grades: Relationship between preschool model and later school success. Early Childhood Research & Practice, 4(1). https://ecrp.illinois.edu/v4n1/marcon.html
9. Minuchin, P., Biber, B., Shapiro, E., & Zimiles, H. (1969). The psychological impact of school experience: A comparative study of nine-year-old children in contrasting schools. New York: Basic Books. ERIC ED047038. https://eric.ed.gov/?id=ED047038
10. Mitchell, L. S. (1921). Here and now story book: Two- to seven-year-olds. New York: E. P. Dutton. https://archive.org/details/herenowstorybook00mitc
11. Nager, N., & Shapiro, E. K. (1999). The developmental-interaction approach to education: Retrospect and prospect. Occasional Paper Series, 1999(1). New York: Bank Street College of Education. https://educate.bankstreet.edu/occasional-paper-series/vol1999/iss1/1/
12. Nager, N., & Shapiro, E. K. (Eds.). (2000). Revisiting a progressive pedagogy: The developmental-interaction approach. Albany, NY: State University of New York Press. ERIC ED440054. https://eric.ed.gov/?id=ED440054
13. Schweinhart, L. J., & Weikart, D. P. (1997). The High/Scope Preschool Curriculum Comparison Study through age 23. Early Childhood Research Quarterly, 12(2), 117–143. https://www.sciencedirect.com/science/article/pii/S0885200697900090
14. Shapiro, E. (1973). Educational evaluation: Rethinking the criteria of competence. The School Review, 81(4). https://doi.org/10.1086/443101
15. Shapiro, E., & Biber, B. (1972). The education of young children: A developmental-interaction approach. Teachers College Record, 74(1). https://doi.org/10.1177/016146817207400101
16. Stebbins, L. B., St. Pierre, R. G., Proper, E. C., Anderson, R. B., & Cerva, T. R. (1977). Education as experimentation: A planned variation model. Volume IV-A: An evaluation of Follow Through. Cambridge, MA: Abt Associates. ERIC ED148490. https://eric.ed.gov/?id=ED148490
17. Stipek, D., Feiler, R., Daniels, D., & Milburn, S. (1995). Effects of different instructional approaches on young children's achievement and motivation. Child Development, 66(1), 209–223. https://cepa.stanford.edu/content/effects-different-instructional-approaches-young-childrens-achievement-and-motivation
18. Yoshikawa, H., Weiland, C., Brooks-Gunn, J., Burchinal, M. R., Espinosa, L. M., Gormley, W. T., Ludwig, J., Magnuson, K. A., Phillips, D., & Zaslow, M. J. (2013). Investing in our future: The evidence base on preschool education. Society for Research in Child Development & Foundation for Child Development. https://www.fcd-us.org/the-evidence-base-on-preschool/
19. Zimiles, H. (1997). Viewing education through a psychological lens: The contributions of Barbara Biber. Child Psychiatry and Human Development, 28(1), 23–31. https://pubmed.ncbi.nlm.nih.gov/9256526/