$1-Billion Question: Do Smaller Classes Work?

Teachers and parents agree that the state’s class-size reduction program has made it easier for students to get individual attention. But here’s the big question: Does it actually boost achievement? The even larger concern: Is it justifying its $1-billion price tag?

Gov. Pete Wilson is pushing to pump an additional $500 million into expanding the 20-student-per-class program in kindergarten through third grades. And recently amended legislation requires school districts that want a share of the funding incentive to conduct an evaluation of the program’s usefulness.

But the state has set up no uniform test or even standards for assessing the program’s effectiveness. And in the absence of such standards, Orange County school districts have set up a hodgepodge of evaluation systems, from comprehensive tests to simple teacher checklists.

Some Orange County districts such as Irvine, Garden Grove and Santa Ana, have not begun to test students, saying that without clear state standards it would be futile to conduct evaluations, which would yield data that could not be compared district to district.

“We have to make sure we’re comparing apples with apples,” Huff said. “If one teacher does it one way, and another teacher does it another way, there’s no way to compare. We have to standardize it somehow so that we will be able to stand up to the scrutiny that has been put on the class-size reduction program.”

Even within a district, how can educators determine whether the new program is working when there is virtually no data from previous years--when classes were larger--for comparison?

It will take a several years to answer that question, administrators said--time during which the state most likely will invest billions more in the program.

The Fullerton School District, for example, this year conducted a series of tests for all its first- through eighth-graders, an unprecedented effort, district officials said. That information will serve as control data for future class-size reduction evaluations. In-depth analyses actually will begin after next year’s test scores come out.

“Will that glean for us valuable information on smaller class size? Probably not,” said Pat Puleo, Fullerton’s director of instructional services. “But through these assessments, we will see in which areas we did well, and which areas we need more development.”

Fullerton’s procedure may be informative to its administrators, but it most likely won’t help the the state much, said Jeremy Finn, an expert on class-size reduction evaluations and a professor of education at the State University of New York. The root of California’s dilemma is its lack of centralized state oversight on the program, he said.

“Your state needs a state-level evaluation in the worst way,” Finn said. “Each district is implementing smaller class sizes in a different way. Asking districts to assess their own program is not enough. If the state is going to do anything really useful, it has to receive the same kind of data from every district.”

Finn was an evaluation consultant for Tennessee’s 12-year-old experiment on smaller class sizes, which was the basis for Wilson’s push for California’s plan. The ongoing Tennessee experiment, which started in 1985, has been lauded by researchers as meaningful evidence that smaller class sizes can help students learn better.

But the Tennessee project doesn’t claim that merely shrinking class sizes will raise student test scores.

“The big question here is under what conditions are the students best served in smaller class sizes?” Finn said.

To determine this, Finn said achievement scores and a multitude of classroom conditions must immediately be accounted for, such as student demographics, the rate of student transiency, the classroom setting and the teacher’s training and level of experience, to name a few.

Such a bank of information would need to be compiled by external researchers or state officials to hold districts fully accountable, he added.

“It’s got to be systematic,” Finn stressed. “It takes people who are into data collecting. Otherwise, you cannot make sense of it afterward.”

Officials of the California Department of Education say they are willing to launch a full-scale study but are at a standstill because there is no money to move forward.

Despite a new law requiring the Legislature to pay the department to hire an independent researcher to evaluate the smaller classes by March 28, 2002, the governor does not intend to earmark that money soon.

“We probably will provide some preliminary funding in the 1998-99 budget,” said Glee Johnson, under-secretary to the governor’s office of child development and education. “I imagine we would begin assessing in the year 2000.”

A full-scale assessment plan would cost up to $1 million.

Because of the funding shortage, the Department of Education is starting a foundation to raise the money, consultant Cathy George said.

Assemblywoman Kerry Mazzoni (D-Novato) has introduced a bill demanding that a research plan be developed by November.

In the interim, the department is sending districts assessment guidelines that stress the importance of data collection, but loosely describe the methods for doing so.

“We are asking schools to collect and maintain this information this year,” George said. “Next year we are going to have prepared regulations.”

She added that the state guidelines are flexible this implementation year because many schools could not fully cut class sizes until the February deadline.

Some Orange County districts, however, said they have not received the state guidelines and criticized the Department of Education for not clearly communicating to them what the legislation mandates.

“We’ve lost our window of opportunity to pretest students [in the beginning of the year] so that we can test them at the end of the year,” Irvine Unified administrator Huff said. “We can do testing. There are all kinds of instruments we can use. But the state is going to have a difficult job of pulling all this data together in a consistent way.”

Districts that already are generating various types of student achievement data include Fullerton, Capistrano Unified and La Habra City. In the first semester, these schools administered a battery of “performance-based” tests, in which students were quizzed individually on spelling, vocabulary, recognition of numbers and alphabet sounds, to name a few. These children will be tested again in coming weeks.

Each district’s assessment plan consists of tests adopted from a menu of about 100 state-approved national exams. The tests may have the mutual goal of measuring a student’s academic growth, but the results are scored on varying scales.

Some educators say the data from these tests can help a teacher diagnose a student’s strengths and weaknesses, but cannot be used to compare how districts are doing.

“It’s used as a classroom aid,” Huff said. “It’s not a standardized test that can help us draw general comparisons.”

In addition, evaluating academic achievements of primary-grade students is complex because young students generally lack the attention span and basic reading skills to undergo standardized tests, educators added.

“First-graders are still in the developmental stages, they have too much to learn before we can test them,” Santa Ana Assistant Supt. Joe Tafoya said.

Dianne Camacho, the Orange County Department of Education’s coordinator of assessment, agreed that standardized exams don’t usually yield statistically useful data for children in kindergarten or first grade.

“Young students have not been taught test-taking skills,” Camacho said. “Standardized tests are tough on them. Also, research has shown that there is little correlation between how students perform on a standardized test in the first grade and how they do in the fifth grade.”

In the absence of a comprehensive state evaluation plan, the county department and districts are meeting monthly to map out their own procedure to be in place by the fall. Camacho said the developing evaluation is intricate because along with this year’s class-size reduction program, schools also have adopted new reading and math curricula and teaching strategies.

“We have to consider so many elements,” Camacho said. “Because if we merely test students and if the ones in smaller class sizes outperform others in previous years, we are not going to know for sure whether it was because of the class-size reduction program or not.”