Three different preschool models operating in one large urban school district were studied to determine the success of each in developing the language, self-help, social, motor, adaptive and pre-academic skills of four-year-olds. A total of 721 randomly selected students in 65 inner-city classrooms were studied. There were no significant differences in teacher qualifications found between the models. All classes exceeded the 10:1 ratio recommended for four-year-olds by the National Association for the Education of Young Children.
In the child-initiated preschool model based on Piagetian concepts of child development, students learn through hands-on activities. They are not taught concepts directly, but learn them through self-directed activities. The teacher facilitates learning by providing children with a wide variety of experiences. She encourages children to choose and plan their activity. She engages them by posing problems and asking questions to stimulate and extend their learning. Children are guided in acquiring skills as needed, and are encouraged to reflect and talk about their experiences.
The academically-directed model is didactic and teacher-led, relying on direct instruction. The lessons are prescriptive and tasks are sequenced. Students’ progress is systematically monitored. Classrooms organized in this way consistently focus on academic skills. Some educators compromise by adopting a middle-of-the-road approach that combines characteristics of both models. This model seeks to provide more structure and direct instruction, while allowing students some hands-on activities. Such programs may stress pre-academic and academic skills but use manipulatives to teach them.
Inner-city children in classrooms where teachers believe in a single, internally coherent theory of how young children learn and develop — either the child-initiated or academically-directed model — did better on standardized tests. Children whose teachers attempted to blend aspects of these diverse approaches performed less well.
Children’s development was not hindered in some areas by a strong academic focus, although their receptive and expressive language and gross motor skills were lower than expected in these classrooms. Children in the child-initiated classrooms learned more basic skills than the academically-directed students.
Benefits especially apparent with African-American children
The benefits of child-initiated learning were especially apparent in African-American children’s performance. There were gender differences as well. Boys performed less well than girls in all three models, but did especially poorly in the combined approach.
Cautions and Conclusions
This field study looked at existing educational practices in urban schools. Although students were chosen randomly, there is no data to prove that classes were comparable prior to beginning school. The large number of students spread across so many schools, however, suggests that these results are representative of this urban population.
These researchers report that policymakers often believe that earlier academic preparation will best prepare young children for school learning. In this study, the academically-directed model was not the worst, but it was not as beneficial as the child-initiated model for these urban children, especially for African-Americans. Also, previous long-term research indicates that emphasis on academics before first grade has negative long-term effects.
The combination approach, sometimes offered as a compromise between two theories of early-childhood education, was found to be the least effective. These classrooms had the lowest basic-skill mastery and were especially detrimental for boys. Classrooms observations revealed that teaching was not as closely aligned with students’ development. Effective preschool curricula matches children’s developmental levels.
“Differential Impact of Preschool Models on Development and Early Learning of Inner-City Children: A Three-Cohort Study”, Developmental Psychology, Volume 35, Number 2, March 1999, pp. 358-375.
Published in ERN May/June 1999 Volume 12 Number 5