Early Ed Watch - logo

New Findings Link the 'Fade-Out' Phenomenon to High-Poverty Schools

You voted. We investigated. In a blog post last month, we asked you to choose what research most piqued your interest among 10 relevant posters released at the biennial meeting of the Society for Research in Child Development. The top 3 vote-getters: Research on "fade-out" in the elementary school years; social behavior in preschool; and early academic outcomes for children in family-based care, center-based or public pre-K. Our final post in this series describes the fade-out research, which is clearly a topic of great interest among our readers, receiving more votes than any other. Thanks again for your input.

Researchers have long puzzled over why poor children who acquire significant cognitive benefits from preschool tend to lose that academic edge by 3rd grade -- a phenomenon known as "fade-out." Research presented last month by Aleksandra Holod and Jeanne Brooks-Gunn of Columbia University's Teacher College fills in another piece of the puzzle, showing that one factor is whether the child's elementary school serves a population that is mostly poor.

The study compared disadvantaged children who attended preschool to disadvantaged children who did not and whose parents were their primary daily caregivers in the years before entering kindergarten. As expected, preschool was linked to higher test scores in reading and math. What's more, the study showed that those benefits persisted most strongly when low-income children went on to attend relatively low-poverty public schools. The impact was not only significant, but quite large, and appeared for both the reading and math scores of the students in poverty.

"Maybe the story here is that, independent of family social class, the kind of school a child goes to makes a difference," said Brooks-Gunn, a professor of child development at Columbia and the co-director of The National Center for Children & Families, where Holod is a graduate fellow. In 2003, Brooks-Gunn testified before Congress that preschool alone was not likely to solve the problems of the achievement gap. Her talk was titled "Do You Believe in Magic?"

For this particular poster, Holod and Brooks-Gunn included children who, according to their parents, attended "preschool" but not Head Start. (The data, unfortunately, did not include any descriptions that might provide a hint of the quality of these preschools.) Brooks-Gunn and Holod said they will soon be doing the same analysis using data on children who attended Head Start or who were placed in informal childcare settings.

Their research strikes some parallels with a 2007 analysis conducted by social science researchers Katherine A. Magnuson, Christopher Ruhm and Jane Waldfogel. Both studies used national data from the Early Childhood Longitudinal Study that tracked children who entered kindergarten in 1998. The 2007 study, published in the Early Childhood Research Quarterly, focused on classroom quality and probed whether the positive effects of preschool persist for longer periods of time if children are given enriching educational experiences in elementary school. They found that the subsequent classroom environment mattered. High-quality environments provided a boost to those without preschool, while low-quality ones hampered the progress of children even if they had preschool experience.

In addition, research from the 1990s by Valerie Lee and Susanna Loeb at the University of Michigan implicated poor-quality middle schools for undermining progress made by Head Start. A later analysis by two economists -- Janet Currie at Columbia and Duncan Thomas of UCLA -- showed that fade-out, particularly for black students, may be related to children attending failing schools with very low test scores. Their study used national data, too, from a sample of children tracked starting in 1988. It connected children's earlier participation in Head Start, as reported by their parents, to their 8th grade outcomes.

In the new analysis, which has yet to be fully vetted and published, Holod and Brooks-Gunn zoomed in on test scores of math and reading skills at two points in the lives of elementary school students: the beginning of kindergarten and the end of 3rd grade.

They examined that data against two variables -- whether children had attended preschool and where they went to elementary school. Elementary school data was broken into three groups, categorized by the socio-economic status of the school's population. If a school's enrollment included fewer than 25 percent low-income children (as determined by their qualification for the federal free and reduced-price lunch plan), the school was considered a high-income or "high SES" school. Schools with between 25 to 50 percent of students in poverty were considered "middle SES" schools and those with more than 50 percent in poverty were tagged "low SES."

The mission was to find answers to a classic "fade-out" question: Were the advantages of preschool more likely to be maintained for students at "less poor" elementary schools?

The first run of the data confirmed what has been found elsewhere: Preschool matters. Attending preschool provided the poorest children with a cognitive boost in kindergarten compared to their poor counterparts who received mostly parental care prior to starting school.

The study showed that if you combine that preschool experience with attendance at a high SES school -- where only 25 percent of children are in poverty -- not only do the benefits last, their impact appears even bigger than at kindergarten. By 3rd grade, for example, poor children in these relatively affluent schools had reading skills that were heads and shoulders above poor children with no preschool who went to poorer schools. (In statistical terms the effect size for reading, for example, was 0.468.)

There is at least one caveat, however: the number of children who fell into this category was small. Of 1,960 children in poverty, only 119 had the experience of preschool plus attendance at a "high SES" elementary school.

On the flip side, and in keeping with the fade-out phenomenon, the effects of preschool diminished greatly for poor children who attended preschool but who went to schools where the majority of children come from poor families.

The research, while still ongoing, already opens many questions. What is it, for example, about the interactions within the walls of a poor school that blunt preschool's effects? "We'll have to figure that out," Brooks-Gunn said. "Is it a resource issue? A training issue? Is it the composition of the classroom, with teachers having more inattentive kids?"

Other research has already pinpointed some troubling characteristics of high-poverty schools that could be to blame. Much of it focuses not on the children but on what their schools have been missing in terms of funding, resources, teacher-recruitment and training.

A 2006 report from The Education Trust, for example, found that teachers in these schools are less skilled and experienced than those serving more advantaged populations. One of its conclusions: "The very children who most need strong teachers are assigned, on average, to teachers with less experience, less education, and less skill than those who teach other children." Education Trust reports from previous years, including one titled "The Funding Gap," have shown that many states and localities scrimp in funding high-poverty districts, providing them with fewer funds per pupil than their more affluent counterparts. And a recent analysis by our counterparts at Ed Money Watch, another New America Foundation blog, shows that funding formulas for Title I programs "do not effectively benefit high-poverty schools, contradicting the program's stated goal."

More evidence of disparities between high- and low-poverty schools can be found in Gordon McInnes's new book, In Plain Sight: Simple, Difficult Lessons from New Jersey's Expensive Effort to Close the Achievement Gap. In it he shows that failing districts are often missing a coherent plan for instruction that allows teachers to assess the needs of their students and tailor their teaching accordingly.

The new fade-out findings provide another platform for continued research into what matters most in improving the quality of children's early education experiences all the way up through the later years of elementary school. "There is work to be done," Brooks-Gunn said, "but from a policy perspective, I think this begins to tell an important story."


This seems to support the

This seems to support the practice of some progressive school districts' use of income-based school assignments to reduce the number of low SES schools. Although the original of intent for many districts was to achieve racial integration, while avoiding the use of racial quotas, the practice does not seem to work well for that purpose. See Reardon, Yun, & Kurlaender. (2006). Implications of income-based school assignment policies for racial school segregation. Educational Evaluation and Policy Analysis, (28)1, 49-75.


Does it ever seem to occur to anyone that there is a significant difference in what goes on in the home of many impoverished families that is the causal agent for "fade"? That possibly poor parents who actively search out schools in higher SES areas are themselves altering the home environment that positively affects their child? Instead of this inceasing focus on schools, perhaps it would be useful for reasearchers to include data on the impact of the family on outcomes. Perhaps "fade" is connected to parental imposition of low expectations of their children - i.e. I failed, your grandparents failed, so don't feel badly if you don't do well in school...the psychological term is called sabotage.


It is true that a coherent and concrete curriculum, backed by support for teachers in their classrooms, is an important element in sustaining and building on the gains of preschool in the primary grades. However, two other points that I try to make in In Plain Sight are more germane: first, that preschool must be HIGH QUALITY, and that the process of making mediocre preschools much better is difficult and takes time; and, second, that the bigger obstacle is that many (most?) city educators do not accept that high-quality preschool is a foundation to improved academic achievement by students from poor families. Instead, they think that learning begins in first grade and all the other stuff before that is just supervised play.