BY MAI MIKSIC | Children start school with varying degrees of “school readiness,” or preparedness for full-time schooling. School readiness includes measures such as children’s socialization level, ability to sit and pay attention, and/or knowledge of the alphabet and numbers. School readiness is also predictive of academic achievement later in life (Duncan et al., 2007), and it is unevenly distributed across children from different socioeconomic backgrounds. On a summary measure of early math and reading skills, learning-related and problem behaviors, and overall physical health, children from low-income backgrounds scored 27 percentage points lower than children from moderate/high income backgrounds (Isaacs, 2012). This troubling statistic indicates that the achievement gap starts early.
Many researchers and policy makers view pre-kindergarten (pre-K) to be a possible equalizer – the impetus behind the recent resurgence in advocacy for universal pre-K. Mayor Bill de Blasio campaigned on and eventually won financial support for New York City’s first universal pre-K program. As other cities (e.g. Seattle) consider implementing their own programs, it’s important to look beyond mere access to universal pre-K and into the quality of funded programs.
One obvious factor is the pre-K curriculum. There are many options from which to choose (Bright Beginnings, Doors to Discovery, Literacy Express, etc.). High/Scope (also known as HighScope) is a popular choice, driven largely by the remarkable success rate of the Perry Preschool program, a small-scale experiment done in the 1960’s which employed the High/Scope model and produced lifelong positive outcomes in a small at-risk African American sample. Today, roughly a quarter (22%) of Head Start programs use High/Scope.
The High/Scope curriculum is based off of four main principles: active learning, positive child-adult interaction, a child-friendly environment, and consistent routines. Teachers encourage children to be independent and learning is highly individualized. Teachers partner with children, providing “scaffolding” to support children’s development, instead of leading the learning experience. Classrooms, organized into “centers,” give children specific areas to focus their attentions. Each center has a theme, such as arts/crafts, science, numbers, or reading. Teachers also foster children’s social development through small or large group interactions.
The Montessori model is another increasingly popular curriculum. Montessori programs differ from High/Scope programs in that they provide mixed-age classrooms where younger and older children can learn from one another. Like High/Scope, Montessori programs are less teacher-directed and more collaborative. The physical space differs from High/Scope in that they are designed to have large open spaces, giving children more freedom to move about as they please and work at their own pace. The Montessori curriculum also has a particular focus on the early development of fine motor skills, with an emphasis on sensory experiences. The Montessori model incorporates children’s cultures into its curriculum, although arguably this is not particularly unique; High/Scope has been known to do the same.
While the High/Scope curriculum has been studied rather extensively, there is less research on the success of the Montessori model. Most research on Montessori programs has been done with White families from high-income backgrounds. Thus, it is relatively unknown whether the Montessori curriculum would be successful with low-income Black or Latino children and more research is required to determine its efficacy. Additionally, there is a need for more research on the High/Scope curriculum because the results have been rather inconsistent; Perry Preschool, while successful, had an extremely small sample size and larger studies of the curriculum have produced disappointing results. A recent study by Ansari and Winsler (2014) adds to the research literature by examining the effects of both the High/Scope and Montessori pre-K curricula on low-income Black and Latino children.
This study used the Miami School Readiness Project (MSRP) as its dataset. Latino (n=7,045) and Black (n=6,700) children were enrolled in Miami’s public pre-K programs. All the children were four years old when they started pre-K and came from low-income backgrounds. Most children (n=12,975) were enrolled in a conventional High/Scope program, while 770 children were enrolled in a Montessori program. In the High/Scope schools sample, 52.6% were Latino and 47.4% were Black. In the Montessori sample, 28.4% were Latino and 71.6% were Black. All eight Montessori schools included in the study were charter/magnet schools. There were no direct measures of children’s parents’ backgrounds; however, since all children came from similar economic backgrounds, low-income status is the best available proxy for other parental characteristics.
The outcomes of interest in this study were children’s cognitive, language, and fine motor skills and their socio-emotional and behavioral problems. The authors used the Learning Accomplishment Profile-Diagnostic (LAP-D) to evaluate children’s cognitive (matching and counting), language (comprehension and naming), and fine motor (writing and manipulation) skills. They chose this test because it aligned with Florida’s Early Learning Performance Standards and was also available in Spanish. They used the Devereux Early Childhood Assessment (DECA) to assess socio-emotional and behavior problems, which consisted of four subscales: initiative, self-control, attachment, and behavioral concerns. Teachers and parents were responsible for observing and recording children’s behaviors using the DECA. The outcomes were assessed at time 1 (T1) in September/October and then again at time 2 (T2) in April/May. The authors thus looked at the change in scores over the course of a school year.
Ansari and Winsler used structural equation modeling (SEM), a rigorous and flexible statistical method used for testing the causal relationships between variables. Although SEM and regular regressions have technical similarities, there are key differences between the two. Only the advantages applicable to the authors’ analysis will be reviewed here. One of the advantages is that it can simultaneously estimate multiple equations. In a regular regression, there is only one outcome variable (the dependent variable). If one is interested in multiple outcomes, then one must conduct multiple separate regressions. SEM can estimate more than one outcome variable at once. This makes SEM a more parsimonious option. Performing several separate regressions individually can not only be cumbersome but may also introduce human error.
SEM is also a good option when comparing multiple groups in multiple domains; in this case Black and Latino children in Montessori programs were compared to Black and Latino children in High/Scope centers. Additionally, SEM is appropriate for data with multiple time points. In sum, the fact that we are dealing with multiple outcomes, multiple groups, and multiple time points indicates that SEM was a good methodological choice.
Using this method, the authors also accounted for the nested nature of the data. Nesting has to be addressed statistically, because children within classrooms (or centers/programs in this case) often share experiences and are exposed to similar variables (such as teachers) that can affect their outcomes. While this study did not have direct measures of teacher background as controls, using intraclass correlation (a measure of how similar children are in a group) in conjunction with the statistical methods in this study can factor out its influences. This is important because the teacher education requirements between the Montessori and High/Scope programs differed. Montessori teachers were required to have a master’s degree, whereas the High/Scope teachers were required to have a bachelor’s degree. No direct measure of this was available; therefore we do not know the extent to which the teacher education levels differed, but it is safe to say that this may have provided children in Montessori programs with an advantage. The statistical methods used in this study were able to take this into account.
All the children exhibited growth in all academic areas. This is the most important finding, and it gestures towards the value of pre-K in preparing young children for school. The strongest differentiated finding is that Latino children benefit greatly from the Montessori curriculum, whereas Black children benefited less from Montessori than from High/Scope.
In the Fall (T1), descriptive statistics showed that children in Montessori schools scored lower on their fine motor skills, exhibited greater behavior problems, and lower social skills compared to children in High/Scope centers. There were no statistically significant differences in children’s cognitive or language skills. The authors controlled for these initial skills to ensure that they did not influence the estimated gains over the course of the years. They then compared the T1 results with those found in the Spring (T2).
The authors’ main conclusion: neither curriculum could be deemed superior across both ethnic groups. All children, as expected, made gains over the time span, but the amount of gain differed between the groups.
The results were reported in terms of effect sizes, which are a standardized measure of the differences between two groups. Traditionally, an effect size of less than 0.2 is considered small to negligible, 0.5 is modest, and anything greater than 0.8 is large. However, it is important to note that effect sizes are generally smaller in education research than in other fields, and these numbers should be interpreted relative to its corresponding number. For example, the effect size for Latino children should be evaluated relative to the effect size for Black children, not evaluated individually for size alone.
For cognitive skills, Latinos in Montessori programs started pre-K with the lowest cognitive scores but made the greatest gains (effect size=0.43-0.52). By the end of the year, the children had caught up to their peers in High/Scope centers. Black children, however, had the least amount of gain over the course of the year. By the end of the year, they were 5-10 percentage points behind their peers in the High/Scope centers (effect size=0.14).
Findings for language skills were similar. Latino children in Montessori schools made the largest gains out of all the groups (effect size=0.22-0.45), while Black children benefited less from the Montessori curriculum (effect size=0.19). In the fine motor skills domain, Latino children in Montessori schools demonstrated, again, the greatest amount of progress (effect size=0.47-0.52); Black children in Montessori schools gained the least (effect size=0.02).
For DECA scores, teacher rated and parent rated skills were reported separately as a way of controlling for the individual parental or teacher biases. Teachers reported that curriculum did not make a difference on social skills, even when ethnicity was taken into account. Neither curriculum was better than the other as far as promoting behavioral skills, but there was an effect when ethnicity was taken into account. Again, Latino children in Montessori schools exhibited the greatest gains in behavioral patterns (effect size=0.19-0.43). Black students in both High/Scope and Montessori curriculums showed in increase in problem behaviors over time (effect size=0.05). Parent rated results were very comparable to the teacher rated results, indicating that there was little bias.
Results showed that Montessori schools positively influenced Latino children’s development. These are promising results, but they lead to additional questions. Why do Latino children do better than Black children in Montessori schools, and what is the mechanism through which this works? The answers are not clear cut.
Ansari and Winsler posit that the emphasis on phonetics in Montessori programs aligns more with the Spanish language than English. Indeed there is evidence that Latino children respond better to phonetic instruction (Cheung & Slavin, 2012). However, High/Scope has also been known to use phonics in their curriculum as well, which calls into question this conjecture.
Second, the authors believe the emphasis on individualized instruction and independent learning could have positive effects on Latino children’s development. Yet, High/Scope also encourages children to learn at their own pace and focuses on individualized learning.
Montessori programs claim to be more focused on incorporating children’s culture into the curriculum. This could help Latino children as they ease into the school environment, which could be significantly different from their home environment. Again, however, High/Scope has also claimed to do the same.
The one key difference between the two curricula is that Montessori classrooms are mixed-age. But it is unclear why one ethnic group would differ from the other in this setting. If the intention is to provide opportunities for older and younger children to learn from each other, it does not follow that one ethnic group would benefit more than the other.
Finally, as stated before, teachers in Montessori classrooms were more educated than those in High/Scope classrooms. But if this were an issue, then Black and Latino children in Montessori schools would have performed relatively the same compared to children in High/Scope. The fact that Montessori Latino children made the most gains out of all the groups (including Black children in Montessori classrooms) indicates that teachers could not have been the only factor influencing their development. Thus it is unclear what specifically about the Montessori curriculum makes it more appropriate for Latino children. Further research is needed in order to substantiate the claims of this study and determine the mechanism for these results.
These results do not mean that one curriculum is better than the other but, rather, suggests that the efficacy of each curriculum may depend upon the ethnicity of the children involved. This is does not imply, however, that children should be separated into different programs based on race or ethnicity The research is far too preliminary to indicate whether one curriculum is truly superior for any ethnic or racial group. Rather, it suggests that it would be inappropriate for policy makers to impose a set curriculum upon pre-K programs, irrespective of the needs and background of the children involved.
This research does, however, support the value of at least these two pre-kindergarten models in preparing young children for school. All the children in this study made sizable gains in cognitive, language, and motor skill development, irrespective of which of the curricula they experienced. It looks like early learning can give children a head start on their formal education.
Ansari, A., & Winsler, A. (2014). Montessori public school pre-K programs and the school readiness of low-income Black and Latino children. Journal of Educational Psychology. Advance online publication. http://dx.doi.org/10.1037/a0036799.
Cheung, A., & Slavin, R. E. (2012). Effective reading programs for Spanish dominant English language learners (ELLs) in the elementary grades: A synthesis of research. Review of Educational Research, 82, 351–395.
Duncan, G. J., Dowsett, C. J., Claessens, A., Magnuson, K., Huston, A. C., Klebanov, P., & et al. (2007). School readiness and later achievement. Developmental Psychology, 43(6), 1428-1446.
Isaacs, J. B. (2012). Starting school at a disadvantage: The school readiness of poor children. Center on Children and Families, Brookings Institute. Retrieved from http://www.brookings.edu/~/media/research/files/papers/2012/3/19%20school%20disadvantage%20isaacs/0319_school_disadvantage_isaacs.pdf