homogeneity and class size

I was speaking with a fellow educator the other day about class sizes and homogeneity. Some thoughts I shared with him....

Class size reduction is an area of thought close to my heart. From my summers working with smaller numbers I have seen that it can have a great effect on my ability to keep in touch with each individual student. Recent studies on the effect of class size reduction do seem to bear this out. In the table below you see the research of three groups[1]. They all compare the increase in learning (in months) to the cost. (It is difficult to measure cost for the second one—thus the ?.)

Interventions

1. Class size reduction from 30-20. 3 month increase in learning in a year at an estimated cost of $30,000.
2. Increase in teacher knowlege from 50th percentile to 94th percentile. 1.5 month increase in learning at a cost of ?.
3. Effective formative assessment. 6-9 months increase in learning per year at an estimated cost of $3000 per year.

Your second point about heterogeneity is one that interests me as well. I went to a school with all homogeneous classes so I don’t have the perspective of learning in a heterogeneous classroom. I do appreciate many aspects of homogeneity and think that combining it with reduced class sizes in specific classes might be the way to go. If 20% of students are failing English 9 then it would seem that one of the possible steps might be to take action by reducing class size to 20-22.

Later in the day I happened to be calling Grant Wiggins to set up a web conference and he talked about the idea whereby competencies might actually lead to “smart” homogeneity. He cautioned that he supports heterogeneity just maybe not in all cases. If I were to extrapolate his thinking I would say that at times homogeneity is the right thing to do and at times heterogeneity is the right thing to do. I know from many of his books and speeches that he is against mindless devotion to any one educational system. He always counsels that as professionals we must thoughtfully decide what is best for students.

My own view is that smart homogeneity is a good idea but one that cannot be done easily. It requires everyone to understand the subtle (all too subtle) differences between homogeneity and tracking. Poorly done homogeneity is more damaging to students than poorly done heterogeneity. That is too sweeping but I feel that there is some truth to it. I support the use of good formative assessment to teach a heterogeneous group in a one room school house style. With good formative assessment you find out who needs what and teach accordingly. Then in the next unit you find out who needs what and teach accordingly. The homogeneous groups within the class are fluid.

On Dec. 3rd I will talk about the third row of the table which is the area where we can take action right now. That said I greatly encourage a push in the community to educate other community member and the school board to the fact that education is an investment NOT an expense. States chose to get into the business of education long ago because they knew it was the right thing to do—even in tough economic times it is important to remember that the best path to future prosperity lies in educating the citizenry.

I had this quote from Dylan Wiliam in another piece I wrote but will add it here as a way of closing:

“If you achieve at a higher level, you live longer, are healthier, and earn more money…In addition, people who earn more money pay more taxes, are less likely to depend on Medicaid or welfare, and are less likely to be in prison. It has been calculated that if a student who drops out of high school would stay to graduate, the benefit to society would be $209,000 (Leve, Belfield, Muenning, & Rousse, 2007). This sum is made up of $139,000 in extra tax revenue, $40,500 savings in public health cost, $26,600 savings in law-enforcement and prison costs and $3000 in welfare savings. Eric Hanushek (2004), a leading economist of education in the United States, has calculated that if we could raise each student’s achievement by one standard deviation (equivalent to raising a student from the 50th to the 84th percentile), over 30 years, the economy would grow by and additional 10%, and just the additional taxes being paid by everyone would more than pay for the whole of K-12 education.”

1. [1] Jepsen and Rivkin (2002)
2. Hill Rowan and Ball (2005
3. Wiliam, Harrison and Black (2004)