September 1, 2006

Size Does Matter

Following up on my last post on reducing class size and why it is a load of bunk, I'll now explain why teaching small groups of students (at least at the elementary school level) is a necessary component in effective instructional programs.

The research tells us that by merely reducing class size, we can expect to see no more than about a 0.25 standard deviation increase in student performance. This is an educationally insignificant result. This unexpected outcome is the result, I believe, of ineffective teaching practices. If what is taking place in the classroom isn't working for the students, then it really doesn't matter whether there are 15, 25 or 50 students in the class. The result will be mostly the same -- students not learning.

But, what if we controlled for teaching effectiveness? What if we only look at highly effective instructional programs? Does class size make a difference in these programs?

The answer, unfortunately, is we don't know for sure. I don't know of any research on this exact point. But, I do know that in the realm of highly effective instructional programs, they all seem to rely on smaller class sizes as a component of the program.

For example, the two highest performing instructional programs, Success For All and Direct Instruction, both rely on small class sizes. I am more familiar with the Direct Instruction program, so I'll focus on that.

Let's first look at the research. Adams summarizes the DI research well:
[T]he meta-analysis shows that 17 studies lasted less than a year and 17 lasted over a year. The effect size can be calculated per comparison and per study but all of the results show large effect sizes: .95 for studies less than a year and .78 for studies more than a year.

[T]he age of the publications was analyzed (1972-–1980: 6 studies, 1981-–1990: 22 studies, 1991-–1996: 6 studies) and all of the effect sizes were large (.73, .87, 1.00, respectively).

I also analyzed the data 8 other ways: by type of student, age/grade of student, academic subjects, test,research design, teacher, fidelity checks, and country. No matter which way the data were analyzed, the results were consistent: implementing DI programs resulted in large effect size gains.

Fifteen of the studies were conducted by researchers who have been somehow connected with Direct Instruction. In contrast, the majority of the studies (18 studies) were conducted by non-DI-connected researchers. The effect size for studies by DI-connected researchers was .99--—a large effect size. The effect size for studies by non-DI-connected researchers was .76--—also a large effect size.
No matter how you slice and dice the data, the result is the same-- a large effect size of about 3/4 of a standard deviation to one full standard deviation. This is four times the effect size we get when merely reducing class size. To put this in perspective, if your typical title I school (performing at the 20th percentile) were to increase its performance by 0.83 standard deviations it would be performing at the 50th percentile (i.e., like an average classroom).

Small class sizes play an important role in DI classrooms. This is because it is necessary for the teacher to be able to identify student errors quickly and provide corrections immediately. Allow me to shift into teacher mode briefly and instead of explaining how this works, let me show you a movie instead. Go here, and click on the movie "How to Set Up a Reading Group Carefully."

Clearly a lot of attention is given to presenting the material to optimally sized groups of students. The rule of thumb is teach lower performers in groups of 8 or less. Higher performers can tolerate larger instructional groups.

Not only is size important, but so is the placement of the students in the class. The lowest performers are placed front and center where the teachers can keep a close eye on them. Higher performs are placed around the outer periphery since they need the least attention. Middle performers are placed in between these two groups.

In later grades more teaching is done to the entire class and the same type of class room set-up is maintained. The optimally sized classroom depends on the learning ability of the students and the teaching ability of the teacher. A good teacher teaching a bunch of high performing kids can tolerate a much bigger classroom than an inexperienced teacher teaching a bunch of low performers.

In both DI and SfA classrooms, close records are maintained by the schools and student progress is monitored constantly. If students are falling behind or a teacher is not performing up to standard, correction can be taken very quickly and a remedy provided. There is no waiting until the end of the school year to recognize that students weren't learning.

These observations only apply to the elementary school level. In later years students are supposed to take on increasing responsibility for their learning. This is made possible in the DI or SfA classroom because most kids will be performing at grade level if they've been through the program in elementary without having significant gaps in their knowledge base.

High schools generally do a decent enough job with students who have been prepared for high school level work. The problem we have to day is that few kids are really prepared for the rigors of high school.

So class size does make a difference in the earliest years of schooling, but teaching effectiveness is a necessary prerequisite for reductions in class size to make a difference in student performance.

No comments: