Guest Opinion | Scott Phelps: A Comparison of PUSD Class Sizes Reveals Significant Differences

As a follow-up to my recent article on how different each of the PUSD schools are in terms of demographics and test scores, I thought I would add a look at class sizes across the PUSD, which show some large differences at the secondary level.

There is generally not easy to understand and easy to access data on the actual class sizes within PUSD, although there are the staffing ratios in the teachers’ union contract that are rough indicators of such sizes. At the secondary level, it is well known that such staffing ratios can be theoretically converted into expected average class sizes, by multiplying by 6/5 or 1.2 given that there are 6 periods in a day but teachers typically have 5 classes of students and a period without a class called their prep period. In the contract, the staffing ratio for middle school classes is 29.5 to 1 and for high school classes it is 29.75 to 1. The corresponding expected average class sizes would be 35.4 for middle school and 35.7 for high school.

The board received enrollment reports last fall which for the first time in recent memory included the class sizes for each period that a middle school or high school teacher taught a class and the class sizes for elementary teachers. This enabled the calculation of an average class size for each secondary teacher. Further, it was then possible to determine the median of these averages for each secondary school, for comparison purposes. That data is shown in the two graphs below, with special education classes removed from the data:

A couple things are immediately noticeable. One is that the class sizes are significantly below the sizes that the contracted staffing ratios would predict. Another is that there are significant differences in the medians of teacher average class sizes across PUSD’s high schools or middle schools.

PUSD’s alternative education programs, the Center for Independent Study (CIS) and Rose City HS, are included in the comparison. These programs have very small median teacher average class sizes. Continuation high schools such as Rose City typically have smaller classes serving students with greater needs. The CIS class sizes were from the fall, and CIS now has a couple hundred more students, so these are probably no longer completely accurate, although CIS has hired more staff since then because of the increase in enrollment. The small teacher class averages also may be a result of many more students enrolling in independent study during Covid and the challenges of staffing the many different grade levels of students and not having enough students at a particular grade level to have a full class size.

Amongst the regular high schools, Marshall has a median teacher average class size that is 43% higher than Blair’s, 26% higher than PHS’s and 23% higher than Muir’s. That means there are significantly more teachers provided per enrolled student for Blair, PHS and Muir than for Marshall. This is because more resources are being provided to staff special programs, e. g., for Blair’s International Academy or its Armenian Academy or its International Baccalaureate Program or its Dual Language Program or its Health Careers Academy or for Muir’s or PHS’s Career Academies or Pathways, etc.

Special programs do require additional staffing because of their unique curricular and other program needs and many schools have such extra staffing for such programs. Unless these special programs are fully enrolled, however, this does result in smaller class sizes for students and teachers at these schools vs. at schools that don’t have a lot of special programs, like Marshall. This is a form of hidden equity in PUSD as the students and teachers get the benefit of those smaller classes. The same pattern can be seen in PUSD middle school class sizes. Sierra Madre Middle’s median teacher average class size is 23% higher than it is for McKinley’s middle school grades, 16% higher than Washington Middle’s (now called Octavia Butler Magnet) and 11% higher than Eliot’s.

For the elementary schools, the data—with special education classes removed—doesn’t show as much variation in median class sizes, as shown below: