text

text

School policy decisions are some of the most highly researched areas of any in the social sciences. Class sizes, to track or not to track, unisex schools, achievement gaps of various types, teaching style, length of the school day, charter schools, technology in the classroom, standardized testing... the list is lengthy and growing all the time. These decisions should be made based on the best available empirical available, not on what's fashionable or fits with educational dogma.

 A major difficulty with doing research in schools is that the effects of various factors are usually confounded. This can be best elucidated with a simple example. Suppose one is interested in studying the effect on performance scores of unisex classrooms. Two classes are selected, one composed of all female students and the other of both sexes. The girls who were in the unisex class do substantially better on standardized tests at the end of the school-year so it is concluded that being taught in a unisex classroom is beneficial for female students. Is such a conclusion justified? The short answer is no, there are all kinds of other factors that may have caused this result. To name just a few: the students in the unisex may have been better learners to begin with than those in the mixed sex class, the teacher assigned to the unisex class may have been better at conveying the the material, the unisex class may have had different (and better) textbooks, the mixed sex class may have been scheduled at a different time of day (like just before lunch), and so on. The technology of designing valid experiments with human beings is very hard and beyound the scope of this discussion. My point is simple, studies that proport to show that such and such is a better way of doing things may not really be valid at all.

There is a vast literature of educational research published in peer reviewed scientifc journals, but it may not be uniform in quality. One needs to be a spohisticated consumer of such information in order to understand the limitations of a particular study or the biases of the authors. While my training has made me a critical reviewer of scientific studies, I also bring to the table my own set of prejudices and preconceptions. I'll try and elucidate the most important ones here. There are, of course, some that are so deeply rooted that I'm not even aware of them, and I apologize in advance for those I've missed.

1. Children are not created equal. They don't bring to the school the same talents, abilities, attention spans, ability for memorization, etc. They differ in their genetic endowment, their socio-economic class, interests, race, sex, involvement of their parents, etc. These factors matter. One size fits all education, or even a few, will not work for everyone.

2. Teachers are not created equal. Most of the factors that effect children also effect their teachers, too. Parental envolvement, not so much. teachers have different styles and abilities. Some are more effective than others. The more effective ones should be rewarded more. Kids can easily figure out who the good teachers are, the administration should make an effort to do so too. There is simply no excuse for an ineffective teacher to be allowed to burrow into the tenure system until retirement. Teachers should be constantly evaluated and should have to re-apply for their own jobs every few years.

3.  Peer groups matter a lot. Kids are desperate to be liked and accepted. They will mold their attitudes and behavior to conform to the group they  belong to. If their friends don't like school, they won't either. The reverse is also true. Peer groups are a major factor in alcohol abuse, smoking and drug use. 

4. The physical surroundings (the building, the books, audiovisual system, etc.) matter only up to a point. If they are really dreadful, they will certainly negatively impact the education process. If they are adequate, as they are in Natick, they are not really a factor (see the "interesting data" tab for more information. Improving them beyond this point (e.g. building a fancy new schhol) may help real estate values by making the town more desirable, but it is money wasted in terms of creating a good learning environment. Spend the money on after school intramural team sports instead. A kid can't be smoking dope if they're playing second base. This is an opportunity to provide positive peer groups as well.

5. Charter schools aren't really schools. They may generate some good ideas, but they cherry pick the best students/most involved parents and thus aren't comparable to a public (open to all) school. They don't scale because they are targeted at the upper 10% or so of learners who have totally different learning styles than an average student.  

6. Remove unnecessary barriers to learning. Unisex classrooms remove the major distraction to learning. Teenage boys are very interested in teenage girls and vice versa. Algebra, not so much. Research is unequivocal on this point, the major beneficiaries are young women. In a unisex class, they are free of the stereotypes that say they should not do so well in math and science. Do it now! Other good ideas: school uniforms (get rid of the highly provocative clothes worn by young women to attract the attention of young men). Longer school day, longer school year, fewer "teacher conference" early releases, more athletics, longer classes (by the time everyone gets settled and started 10 minutes of classtime have been eaten up). Focus on the basics - teach history, reading, science and technology, applied math. Most kids are not going to Harvard but they need to become good, informed citizens.

7. Make it easier for seasoned professionals to become teachers. The current certification process protects those who have jumped the appropriate hoops. An engineer or accountant or soldier at the end of their career has a lot of real-world experience that kids could benefit from. Allow the miltary to recruit in the school. They are a revered profession, the people who protect us and keep us free. Seniors should be encouraged to consider the military as a carreer. It's the best path to good citizenship that I can think of.

8. We are on the verge of a major transformation of education process due to a confluance of several enabling technologies. Currently, educational practice has more in common with the one room schoolhouse of the 19th century than what it may become in just a few short years. In perhaps 20 years or less, education may become individualized, with highly interactive, intelligent software that provides each learner with material suited to their individual pace and style. It may also become distributed, with the idea of a monolitic centralized school entirely unnecessary. Some of the requisite enabling technologies may already be familiar to you, such as the internet with search engines such as GOOGLE. Others, such as computer adaptive testing, already used widely in such tests as the Graduate Record Exam, may become generalized to computer adaptive learning, where the specific content and pace is determined by immediate feedback from the learner. What this will mean for our current cadre of teachers is unclear; some will adapt, others may move out of the profession. These changes are inevitable, as states and school districts are squeezed monitarily, education must take advantage of the economies of automation. We will no longer be able to afford the "teacher in every classroom" mode of operation.

9. There are probably other things that I can't think of at the moment, I'll add to this web site as they occur to me.