The Relationship Between Smaller Class Sizes and Better Student Outcomes

In recent articles, we have pointed out that Washington state is 46th in the nation in school funding as a percent of income. Washington state is also 46th in the nation in class sizes. It would take several billion dollars per year in additional funding to restore Washington state to the national average in school funding. One of the goals of restoring national average school funding is that it would allow us to hire tens of thousands of teachers needed to lower class sizes in Washington state down to the national average.

Sadly, there are many misleading claims made by those who oppose restoring school funding to the national average. For example, some claim that class sizes are not that high in Washington state. Others claim that reducing class sizes would not help improve student outcomes. In this article, we will clarify how to properly measure class size. We will then compare class sizes to other states and other nations. Finally, we will summarize the research on the significant long term benefits of lower class sizes.


Why Actual Class Sizes are Much Larger Than Student to Teacher Ratios
The most common mistake made when discussing class size is to confuse class sizes with Student to Teacher Ratios. The Student to Reacher Ratio is determined by dividing the total number of students in a school or a state by the total number of professional staff at the school or the state. For example, if you go to the Washington State OSPI website and click on Apportionment, then Publications, then Personnel Summary Reports, then select a year, then click on Table 46, you will get a report called “Ratio of Students to Classrooms.” This is actually the Student to Teacher Ratio. For the 2014 school year, this ratio was 18.2 students per teacher.

This type of statistic might mislead one into believing that the class sizes in Washington state are only 18 students – which would mean Washington state has the lowest class sizes in the nation and in the world. Yet if you walk into any real classroom at any real school in Washington state and count the actual students, you will see about 30 students in the real classroom. Many classrooms will have 35 to even 40 students! Why is there such a huge difference between the Student to Teacher ratio reported by OSPI and the number of students in real classrooms? The problem is that OSPI uses an extremely broad definition for classroom teacher. Many so-called classroom teachers are actually administrators and counselors. Using Student to Teacher ratios therefore misleads the public and even legislators into thinking that class sizes are not that bad when in fact class sizes in Washington state are among the highest in the nation.

A better estimate of classroom teachers comes from a national survey of classroom teachers in which teachers are asked how many students are in their average classroom. This survey indicates that for Grades 1 through 6, the national average class size is 21 students and the average class size in Washington state is 24 students. For Grades 7 through 12, the national average class size is 27 students and the average class size in Washington state is 30 students.

Here is a distribution of class sizes showing which states have low, average, above average or extremely high class sizes:


However, even this survey of teachers under-reports the actual class sizes in the nation and in Washington state because it includes Special Education teachers who often have classes of under 10 students. Excluding Special Education classes, the typical or median class size in the US is likely close to 29 students and in Washington state, it is likely close to 32 students.

Even within Washington state, there is a wide range of school funding and class sizes. For example, the Snoqualmie Valley School District in East King County is funded at a rate that is 10% below the state average. This results in class sizes in the Snoqualmie Valley School District being about 10% above the state average. It is not uncommon to have classes of 35 to 40 students – with some classes climbing as high as 100 students!

If the Class Size Initiative 1351 were fully funded, it would lower class sizes in Washington state down to 17 students in elementary school and 25 students in high school. However, even this would still be considered a high class size in many developed nations. For example, in Finland, the average class size for both elementary and secondary school is 20 students per class.

Why are there conflicting claims about the benefits of small class sizes?
It should seem obvious that student to teacher interaction would improve with smaller class sizes. This benefit show be most obvious with struggling students who would be more likely to get the help they need to catch up with their peers.

Yet despite the importance of this topic, there has been only one large scale double blind scientific study comparing smaller to larger classes. This was the STAR study involving nearly 12,000 students in Tennessee from 1985 to 1989. Elementary school students at 79 elementary schools were randomly assigned to either a small class of 13 to 17 students or a normal size class of 22 to 26 students beginning in Kindergarten and continuing through 3rd Grade. This study confirmed that smaller classes helped students and especially low income and minority students in a variety of important ways with benefits that lasted all the way through high school and into adult life. We will discuss this study in a moment.

Sadly, numerous so-called experts have the audacity to claim that small classes would not help students! For example, a report for the Washington state legislature in 2007 concluded that smaller class sizes were not worth the added expense above the 3rd Grade.

The problem with this study is that it gave equal weight to 38 studies – none of which were large scale random studies. In addition, the metric being used to determine whether smaller classes were effective was high stakes test scores. These test scores are based on tests that are known to be unfair, unreliable, invalid and a complete waste of time and money. In sort, the conclusion that small class sizes does not matter is based on junk science. Other educational outcomes such as student engagement, teacher grades and graduation rates are much more reliable than test scores.


Academic Benefits of Smaller Class Sizes
Smaller classes meant that teachers were better able to give individual attention o struggling students. Students in smaller classes learned to engage confidently with their teachers in asking questions to get help they needed. This success and confidence carried all the way through high school – many years after the experiment in smaller classes ended and the students were returned to normal classes.

As the following chart shows, low income students who were lucky enough to have four full years of smaller classes were much more likely to graduate than their peers who only had zero to three years in the smaller class sizes:


Source: Finn, J. D., et. al. (2005). Small Classes in the Early Grades, Academic Achievement, and Graduating From High SchoolJournal of Educational Psychology.

Economic Benefits of Smaller Class Sizes
A 2011 study summarizing the life academic and economic outcomes of students in smaller classes in the STAR study compared to their peers who had normal class sizes found that “The effects of class quality fade out on test scores in later grades but gains in non-cognitive measures persist.” Put in plain English, high stakes test scores are not an accurate predictor of future student performance. However, student engagement is. Here are just some of the adult outcomes for these students 20 years later of being in a smaller class in elementary school: Students were significantly more likely to graduate from high school, attend college, start a savings account, buy a home, get married and stay married. Students were less likely to commit a crime or go to prison. Much of this information was obtained from federal tax returns of 95% of the nearly 12,000 students involved in the study.
Source: Chetty, R., Friedman, J.N., Hilger, N., Saez, E., Schanzenbach, D.W., & Yagan D. (2011). How does your kindergarten classroom affect your earnings? Evidence from Project STAR. Quarterly Journal of Economics, 126(4), 1593-1660.

In a separate analysis, Alan Krueger, Chair of the Council of Economic Advisers, estimated that every dollar invested in reducing class sizes yielded about $2 in long term economic benefits.


Conclusion... Smaller Classes Lead to More Successful Students
Wealthy private schools understand the importance of small class size. For example, at Lakeside Private School in Seattle, average class sizes are 16 students.

If class sizes of 16 students is considered ideal for the children of the wealthy, these class sizes should also be available to all students in Washington state.


Given the huge academic and economic benefits of smaller class sizes, it is appalling that the Washington state legislature continues to deprive our public schools of adequate funding and deprive our students of the reasonable class sizes they need and deserve. In does not have to be this way. Senate Bill 6093 would increase school funding above the national average and lower class sizes below the national average merely by repealing a 1997 tax break used by billionaires to avoid paying their fair share of state taxes. In the next article, we will review how many new schools and classrooms would be needed to restore national average class sizes here in Washington state. We will then look at how to build a grassroots campaign for passing Senate Bill 6093 here in Washington state.

David Spring M. Ed.
Coalition to Protect our Public Schools