In 2006, Johns Hopkins University sociologist Karl Alexander and his colleagues concluded a study that had followed Baltimore school children for 24 years. The resulting mass of data from these students shed considerable light on the achievement gap between low-income students and their better-off peers, including the effect we now refer to as "summer slide." We recently asked Dr. Alexander five questions about the findings.

1) What did you discover from the Baltimore City study you began in 1982?

Back in the fall of 1982, my colleagues and I began tracking the educational progress of just under 800 Baltimore school children who were then starting first grade in 20 of the city's public elementary schools. It turned out to be a project with a long lifeline, as it didn't conclude until 2006 when we interviewed 80 percent of the original group at ages 28 to 29. Along the way, we had compiled a considerable data archive on these youngsters, their families, and their school experiences, including achievement test scores from school records.

At the time, the city school system was testing twice annually, fall and spring, and because we were following the same children over time, we were able to ask questions about the group's academic development. One line of research has been to compare summer learning patterns against school year learning, following a project reported some years earlier by Barbara Heyns in Atlanta ("Summer Learning and the Effects of Schooling.") Barbara found that achievement differences across social lines - by race and family income - widened over the summer months, more so than during the school year, leading her to conclude that schools helped compensate for out-of-school family disadvantage.

We thought this an important insight and decided to explore the issue for our children. Heyns examined children in middle school and her data covered just one summer tucked between two school years. We looked at an earlier period of schooling, starting at first grade, and over a longer period of time: the five years of elementary school and the four summers they bracket. For us, foundational skills were at issue - reading comprehension and math concepts, the building blocks for all later learning.

What we found is that lower-income children start first grade already behind, and over time the achievement gap between them and children from more advantaged families widened quite considerably. In reading comprehension, the gap went from roughly a half grade equivalent at the outset to three grade equivalents at the end of fifth grade.

That in itself is distressing, but hardly surprising. We know from many studies that the achievement gap tends to widen over time. What was new, and seen for the very first time in our studies, was that almost all of the increase in the achievement gap traced to differential summer learning – disadvantaged children pretty well kept pace during the school year, but fell back during the summer months.

During the summer, children from better-off families continued to build up their skill set – they scored higher, on average, at the start of the new school, than they had scored at the end of the previous one. But this was not so for lower-income children, who essentially treaded water, averaging about the same at the start of the new school year, increasing a point or two or decreasing a point or two from summer to summer. The differences accumulate across summers, and so over time the summer learning differential is quite substantial.

This is the pattern we came to call “summer slide” or “summer setback,” and because we were following the same children over many years, we were able to look also at consequences down the road. Specifically, we documented that two-thirds of the achievement gap across social lines that exist in ninth grade, could be traced back to differential summer learning during the elementary school years. And quite tangible consequences followed, including a lesser likelihood of placement in the college-bound high school track, elevated risk of dropout, and reduced prospects for attending college.

Over the years since, other studies, national and in other localities, have documented a similar summer slide, so it is fairly secure that the long summer layoff is an impediment to disadvantaged children’s learning and ability to keep up. However, these other studies all are short-term - just one summer - and do not assess longer-term repercussions. This pattern of “summer slide” is a profoundly important insight, and it is gratifying that we have been able to contribute usefully to this work.

2) What's different about the summers of low-income children from their higher-income peers?

From surveys of our study children and their parents, we know that family privilege shows up in a whole host of ways, including the obvious enriching experiences. During the summer, they more often go to museums, take lessons of various sorts, get involved in organized sports, and - importantly - go to the library. They have more books in the home, their parents read to them, etc.

But family privilege isn’t reducible to a checklist of activities or the availability of educational toys. All our families had children in Baltimore’s public schools, so these “better off” aren’t the wealthiest of the wealthy. Even so, differences across social lines are large and consequential. Disadvantaged parents hadn’t finished high school; the parents in better-off families averaged some college. It is expecting a great deal of poor parents who have struggled themselves at school to nurture their children’s intellectual growth in the same ways and with the same results as college-educated parents.

Let me add immediately that these parents are no less caring – they want the very same things for their children that the rest of us want, but they lack the means. Heyns’s study concluded that schools help fill that need and our research indicates much the same.

3) At what age does the summer slide start?

Our research starts in the fall of first grade, but national data show much the same pattern for the summer between kindergarten and first grade, so the pattern is discernible from the very outset.

But it’s not just a matter of summer learning contrasting with school year learning. We can detect the pattern then because that’s when we start to test. But, in fact, studies of language development show differences – large ones – in the language richness of the home environment back to birth.

One study – small scale, but compelling – established a cumulative difference across social lines in word exposure in the millions by age three. And it’s not just the total number of words uttered, but differences in complexity, nuance, etc.

The fact of the matter is that children are in their families all of the time, but in school only part of the time, which makes family disadvantage not just potent, but an ongoing process all throughout children’s development.

4) Should these kids attend year-round school?

Two advocacy groups love this research - advocates of year-round schooling and advocates for expanded summer programming. There is also a strong after-school movement. I think the boundaries are starting to blur, as the larger construction of the issue is expanded learning time for needy children. Quality programming should include a rigorous academic component - that much is obvious - but also enriching experiences of the sort middle class children can take for granted, nutritious meals, and a safe and secure setting.

I tend to favor high quality summer programming, as that lends itself to greater experimentation. So, in the ideal, we would learn about “best practices” by comparing the effectiveness of different models. I don’t feel strongly about year-round schooling, but I do favor an approach that adds days to the school year. Most modified school year calendars don’t do that – they take the same 180 to 182 days and chunk it differently.

5) What can be done by society to combat the summer slide?

In general terms, it would be to provide more high-quality learning opportunities for needy children. That could be in schools, as in summer programs, but also libraries, Boys and Girls Clubs, scouting, mentoring programs, you name it. There is a lot of interest nowadays in better summer programming, and that’s great, but it needs to be sustained and ongoing.

Summer school is still often thought of as a luxury, and when budgets tighten, summer programming gets hit hard. You can see that today – with programs being cut back and eliminated all over the country. But children’s needs don’t come and go with the condition of the economy. Disadvantaged children and their parents need our help, and to really help, it must be sustained.

We’ve made progress - at least we’re now talking about these issues – but a great deal more needs to be done.

All statements and opinions expressed on this blog are those of the individual contributors, and not of the Bill & Melinda Gates Foundation or NBC News.