K-5 Class Sizes in Milwaukee Public Schools

Analysis by Russ Allen, WEAC Instruction and Professional Development

On April 30, 1998, a two-page questionnaire was sent to nearly 2,000 elementary school teachers in the Milwaukee Public Schools by the Milwaukee Teachers’ Education Association. The questionnaire was designed to collect information about class sizes and related issues from teachers of self-contained classrooms, kindergarten through fifth grade.

It needs to be understood that the Milwaukee Public School System has class size guidelines for elementary and K-8 schools. For example, kindergarten and first grade classes have a 25 student guideline. The guidelines for grades two through five are twenty seven students per teacher. Except for three- and four-year old kindergarten, the guidelines can be exceeded.

Excluded from this study were three categories of teachers: (1) teachers in the Student Achievement Guarantee in Education (SAGE) Program (whose classes consist of 15 or fewer students per teacher), (2) teachers whose classes consist entirely of Exceptional Educational Needs (EEN) students, and (3) teachers who meet with multiple groups of students during the school day (e.g., teachers of art, music, physical education, or similar subjects).

In all, 779 questionnaires were returned (39% return rate). The final data set consists of 732 respondents.

As noted, teachers who instruct only students with Exceptional Educational Needs are excluded from this study. Nonetheless, most teachers who returned a questionnaire indicated that their classes include several students who are classified as EEN or ESL (English as a Second Language).

Results for Milwaukee Public Schools

The numbers of teachers per grade level who returned the questionnaire are shown in Table 1.

Kindergarten Teachers

The 181 kindergarten teachers who participated in this study were asked to describe their teaching situation and indicate the number of students in their classes.

Teaching Assignment

Seventy-two percent of kindergarten teachers report that they teach all-day kindergarten. Nearly all others teach different groups in the morning and afternoon. Two persons indicated they were part-time teachers.

Number of Students in Kindergarten Classes

The typical all-day kindergarten class in MPS has 25 students, which is in line with MPS guidelines for students of this age. (Note that 25 students is the mean, median, and mode for all-day kindergarten classes. For morning and afternoon classes only, the median and mode are 25; the mean is 23.2).

For the state as a whole, most kindergarten teachers report approximately 21 students per class (mean = 21.7, median = 20/21, mode = 20). This means that the typical kindergarten class in MPS has four or five more students than schools elsewhere in Wisconsin.

Self-Contained Classrooms, Grades 1-5

Five hundred and thirty-five MPS teachers who participated in this study say that they teach the same group of first, second, third, fourth, or fifth grade students for the entire day.

For MPS teachers, the typical class has 27 students (mean = 27, median = 27, mode = 25). These figures comply with MPS guidelines for students in grades 1-5. In comparison, the average first through fifth grade, self-contained classroom throughout Wisconsin has 22 students.

Compared with the rest of Wisconsin, MPS first through fifth grade classrooms have an average of five more students (23% larger).

Changes in Class Sizes

Twenty-six percent of the teachers who participated in this study say that class sizes have increased over the past two or three years. As for the remainder, 9% of MPS teachers say class sizes have decreased; while two-thirds say they have stayed the same.

Statewide, 27% of teachers say class sizes have increased, while 23% of teachers report that class sizes have decreased (compared with 9% in MPS). One-half say they are about the same.

Class Size Preferences

When asked what the maximum class size should be for the students they teach, kindergarten teachers say the maximum class size should be about twenty students . As for the ideal class size, the typical kindergarten teacher prefers 15 -16 students.

Teachers in grades 1-5 say the maximum class size should be no greater than twenty or twenty-one, while the ideal class size is perceived to be between 17 and 18 students.

Meeting the Needs of Each Student

A related question asked teachers to think about the extent to which they are able to meet the individual needs of students, given the number of students currently taught.

Ninety percent of respondents said they have too many students, whereas 8% said the number was about “right.” Two percent said that in some classes they have too many students, while in others the number is about right. Not one person out of 732 respondents said he or she had too few students.

Classroom Space

The final set of questions asked about the use of “non-classroom” spaces for instruction. This is an important issue because in many buildings reduced class sizes would only exacerbate existing space problems.

The percent of teachers saying that their building makes use of other locations for “classroom space” is shown in Table 4. It is significant that 64% of respondents indicated that in their school at least one of the non-traditional classrooms listed in Table 4 is used for instruction.

Concluding Remarks and Significant Findings

A study such as this which deals almost exclusively with numbers can lead one to overlook the human dimension of what is being reported. Although averages, percents, and median values are important in understanding the conditions which exist in Milwaukee Public Schools, it is important that these data be placed within the context of what has happened to Milwaukee since the 1970s and 1980s.

In the 1998 “Quality Counts,” Education Week devotes considerable attention to the conditions of Milwaukee and Milwaukee Public Schools. To this end, they quote Myron Orfield, an expert on urban planning and author of Metropolitics: A Regional Agenda for Community and Stability, who describes Milwaukee as “one of the most distressed cities in America. . . “ (p. 267). Education Week also reminds the reader that Milwaukee’s growth in child poverty in recent years has outpaced any other big city in the United States.

There are five findings in this study which are particularly noteworthy. As they are considered, one must not ignore the existing socio-economic conditions of Milwaukee or the solid body of research which shows that smaller classes can have significant and lasting positive effects on the achievement levels of children.

1. Responses of teachers who participated in this study confirm that MPS guidelines for class sizes in grades kindergarten through fifth grade are being met. However, these guidelines far exceed the optimal class sizes identified by the STAR study (15-17 students)
2. Compared with the state of Wisconsin, MPS classes in grades K-5 are significantly higher. On average, MPS kindergarten classes have four to five more students, while in grades one through five, classes in MPS average five additional students.
3. When asked about the maximum and ideal class sizes, participants in this study indicated that the maximum number should be no greater than about twenty students. The ideal class size was 15-16 students among kindergarten teachers and 17-18 students for the other teachers. Few classes in MPS are this small.
4. In response to the question, “In terms of meeting the individual needs of all students, how do you feel about the number of students you currently teach?”, nine in ten teachers said they had too many students. Not one of 732 respondents said he or she had too few students.
5. Nearly two-thirds of respondents indicated that their school is using at least one non-classroom space for instruction. This means that reduction in class sizes requires more than just changing the guidelines and hiring more staff. In many schools, adequate space is lacking.

Supporting Research

Prior to 1985, research on the achievement benefits of reduced class size had produced mixed results. Although the common-sense conclusion of most educators was that smaller classes would result in more effective teaching and higher student achievement, historic class size research was not unequivocal in support of small class sizes. Class size research prior to 1985 tended to use small numbers of students, generally used a limited time frame (a semester or less), and usually did not establish appropriate control groups.

Project STAR

In 1985, the state of Tennessee, under the leadership of Governor Lamar Alexander, funded Project STAR (Student-Teacher Achievement Ratio). The project authorized a four-year study of the effects of class sizes of 15-17 students in grades K-3 on student achievement. Project STAR included 6,500 K-3 students in 320 classrooms in 80 buildings located in inner-city, suburban, urban and rural settings. Each school building included at least one classroom for each of the three treatments tested: (1) a class of 15-17 students with a regular teacher, (2) a regular-sized class of 22-25 students with a regular teacher, and (3) a class of 22-25 students with a regular teacher and a teacher’s aide.

Standardized and curriculum-based tests in reading, mathematics, and basic study skills revealed that after four years the small class treatment produced substantial improvement in these areas as compared with the other two treatments (Mosteller, 1995). Students in all settings- suburban, urban, inner-city, and rural- benefited from the smaller classes. The presence of a teacher’s aide in a regular-sized classroom did not significantly affect student achievement. Though aides certainly assist teachers, the effect on student achievement is negligible.

Lasting Benefits Study

Beginning in 1989, Tennessee began to track 4,300 students who had participated in Project STAR to determine whether the gains made by students in the small classes would be retained after they returned to regular-sized classes in the 4th grade and beyond. The Lasting Benefits Study (1989-present) indicates that children who had originally been in small K-3 classes continue to score higher than those who had been in regular-sized classes. The “small class” gains have not washed out over time. The students of Project STAR have been tracked through grade eight, and the differences in achievement between students who studied in large and small class sizes remain significant. The powerful continuing effect of learning in a small class in grades K-3 has surprised many observers. Tennessee continues to track students from all three treatments in Project STAR as they move through their high school years.

Project Challenge

In 1989, based on the impressive achievement results attained in Project Star, Tennessee moved to reduce class sizes in grades K-3 in the seventeen economically poorest districts in the state (Project Challenge). Between 1989 and 1993, these seventeen districts improved their end-of-year rank among Tennessee’s 139 districts from well below average to above average in reading and mathematics.

In summary, the Tennessee projects have produced compelling evidence that smaller class sizes (15-17 students per classroom) in grades K-3 result in persisting achievement advantages for students throughout their K-9 education.

Educational Testing Service Study

In 1997, Educational Testing Service published a research study by Harold Wenglinsky in which he studied the relationship between spending and student achievement by analyzing data from three separate sources: National Assessment of Education Progress, the Common Core of Data, and the Teacher’s Cost Index of the National Center for Educational Statistics.

Wenglinsky’s research shows that increased spending for smaller classes has a direct positive effect on student achievement for fourth grade students. Among eighth graders, the relationship is somewhat more complex, in that reduced class sizes improve the social environment (leading to fewer problems) which improves student achievement.

References

Mosteller, Frederick. “The Tennessee Study of Class Size in the Early Grades.” Critical Issues for Children and Youth 5 (Summer/Fall 1995): 113-127.

“Quality Counts: The Urban Challenge. Public Education in the 50 States.” Education Week Supplement, 8 January 1998.

Wenglinsky, Harold. When Money Matters. Princeton, New Jersey: Educational Testing Service, 1997.

Posted June 9, 1998