When Something Does Not Work

Multiple factors determine student performance and some of these factors are key. Good health, for instance, is expected to be an important factor. With this in mind, a good night sleep is worth our attention. Sleep not only allows for a body to rest, but also prepares the brain for the next day. Irritability and difficulty in paying attention are among the common results of sleep deficiency. Sleep deprivation is therefore a possible hindrance to good learning. It is thus expected that a positive correlation exists between having adequate quality sleep and performance in school. For this reason, school start times have become a variable that one may tweak to help improve student performance. Starting school very early in the morning forces a child to wake up early, reducing the number of hours a child could possibly sleep. Numerous studies have shown that later school times can indeed improve student performance.

Of course, it is still possible to see scenarios where what is expected from published studies do not materialize. Education is multivariate. There are multiple factors and in some cases, there is a factor that simply overwhelms the rest. A recent paper, "Earlier School Start Times as a Risk Factor for PoorSchool Performance: An Examination of Public Elementary Schools in the Commonwealth of Kentucky", published in the Journal of Educational Psychology illustrates this concretely.  This study looks specifically at the effects of later school start times on student performance. As in other studies, a correlation between later times and student performance is seen. Students perform better across all subjects when school starts about an hour later. Apparently, this also applies to elementary school children and not just adolescents. This, however, is not the surprising finding. What is remarkable is that delayed school start times improve student performance only in schools where the majority of students does not come from poor families. A later start for school does not have any effect on poor students. This is summarized in the following figure:

Graph above drawn based on data from "Earlier School Start Times as a Risk Factor for PoorSchool Performance: An Examination of Public Elementary Schools in the Commonwealth of Kentucky"
Schools in this study have been categorized as "not poor" or "poor" depending on the number of students who qualify for a free lunch. Considering data for free lunch eligible students in Kentucky, "not poor" are schools that have about a third of its students qualifying for free lunch while "poor schools" are those that have as much as ninety percent of students qualifying for free lunch. Later school times clearly correlate with performance across all subjects in schools not overwhelmed by poverty. On the other hand, in schools where poverty is evident, later school times do not have any effect.

The lesson here is that sometimes something expected to work does not work and the reason is that there is a different factor that is overarching. In this case, it is poverty.