Academic achievement and between-class transition time for self-contained and departmental upper-elementary classes.

The study investigated the relationship between elementary school classroom organizational structure (i.e., self-contained versus departmental formats) and standardized achievement scores, transition time between classes, and instruction time. Participants included 103 fifth-grade and 94 sixth-grade students from one school district. Based on previous findings, students from self-contained classes were predicted to achieve significantly more than comparable students from departmentalized classes, take significantly less time to change classes, and spend more time in instruction. Results indicated that the self-contained group gained significantly more on Total Battery, Language, and Science subtests compared to the departmentalized group. Departmentalized classes took significantly longer to transition from subject to subject than did the self-contained classes. No differences were evident for instructional time. Findings were consistent for fifth and sixth grades. The results are limited because of only using one school district.

**********

Educators have debated elementary school organizational structure since the beginning of the twentieth century (Gibb & Matala, 1962; Lamme, 1976). One aspect of organizational structure involves the number of subject areas covered by each teacher. In the self-contained approach, the teacher acts as a generalist and carries responsibility for the curriculum all day. The other extreme is the departmentalized approach. Here students change teachers for instruction in different subjects. Thus teachers cover fewer subject areas (Roger & Palardy, 1987; Mac Iver & Epstein, 1992). Advocates for a self-contained organizational pattern argue that it promotes instruction which is child-centered rather than subject-centered. Self-contained classrooms allow the teacher and students the opportunity to become well-acquainted. Moreover, self-contained teachers know their students' strengths, weaknesses, and personality traits enabling better accommodation of the students' individual learning styles (Squires, Huitt, & Segars, 1983). Additionally, self-contained classes allow for greater flexibility in scheduling. Elkind (1988) argues that the time students spend gathering books and paper and moving to other departmental classes cuts into valuable instruction time.

On the other hand, some educators have found that departmentalized organizational approaches offer distinct advantages for the student (e.g., Culyer, 1984). Anderson (1962) presented a strong case for specialization when he reported that only 4 of 260 teachers considered themselves well prepared in all subject areas. Walker (1990) noted greater emphasis on curriculum matters in departmentalized elementary schools.

This paper is similar to one by Garner and Rust (1992) which found that fifth-grade students in self-contained rooms scored significantly higher on group achievement tests compared to their departmentalized peers. The present study added measures of transition time and actual instruction time.

Method

Participants

The participants included 197 students (103 fifth graders and 94 sixth graders) from two kindergarten-sixth grade schools in rural Tennessee. There were 109 students from school A. Of these students, 58 fifth graders (30 boys and 28 girls) and 51 sixth graders (23 boys and 28 girls) attended departmentalized classes. School B's participants included 88 students. Of these students, 45 fifth graders (20 boys and 25 girls) and 43 sixth graders (21 boys and 22 girls) attended self-contained classrooms. All of the participants attended self-contained classrooms in the fourth grade. School A used departmentalized fifth-and-sixth grade classes. School B maintained self-contained classes through Grade 6. The social class compositions of Schools A and B were similar, with 27% of the students at each school getting free or reduced-fee lunch.

Apparatus

The scale scores and normal curve equivalents of the norm referenced component of the Tennessee Comprehensive Assessment Program (TCAP) were dependent variables. The primary aim of the instrument is to provide a measure of achievement of basic skills in reading, spelling, language, mathematics, study skills, science, and social studies. The Tennessee Department of Education considers the scale scores obtained on the TCAP useful for measuring growth of students or groups of students from year to year (Tennessee Comprehensive Assessment Program, 1993). Scores range from 0 to 999. Parents, students, and teachers can monitor annual progress up the scale in each subject from 0 toward 999. McWherter and Smith (1993) refer to year-to-year scale score comparisons as value-added assessment. The TCAP also supplies traditional normal curve equivalents (NCEs) which have many of the characteristics of percentile ranks, but have the additional advantage of being based on an equal-interval scale. NCEs have a mean of 50 and a standard deviation of 10 (Bock & Wolfe, 1996).

Transition time was recorded by direct observation. Each group was observed for two full days, not consecutively and not the same day of the week. Actual time was recorded and rounded to the nearest minute from the closing of one subject until the beginning of the next. Instruction time was recorded during the same days. Subject matter was noted. The first author made these observations.

Results

Normal curve equivalents (NCEs) and scale scores from the TCAP Total Battery and subtests (Reading, Language, Mathematics, Science, and Social Studies) were analyzed in six separate 2 (Grades: 5, 6) x 2 (Organizational Structure: Self-contained and Departmental) x 2 (Male, Female) analyses of variance. The mean gains were calculated by subtracting the difference between TCAP pretest scale scores from posttest scale scores (see Table 1). School A implemented a departmentalized organizational structure, and School B implemented a self-contained structure.

Significant effects were found for gain scores using the TCAP scale scores. Self-contained students gained significantly more than departmentalized students in Total Battery, Language, and Science in fifth and sixth grades (see Table 1). No differences were found in Reading, Mathematics or Social Studies. Inspection of Table 1 reveals wide differences in gain scores. One of the 5th grade classes lost points in science. All of the other groups improved compared to the previous year's scores. The gain scores found here are more variable and more modest than those presented in a longitudinal study of the TCAP (Bock & Wolfe, 1996).

In the NCE analyses, there were no significant main effects or interactions for organizational structure. Transition time was significantly more efficient in the self-contained classes compared to the departmentalized school (average transition time was 3.27 minutes for the self-contained groups compared to 4.55 minutes for the departmental groups). However, there was no significant difference in actual instruction time. Departmentalized classes averaged 48 minutes of instruction time per hour while self-contained classes were engaged in instruction an average of 46 minutes. That difference was not significant. Anecdotal observations revealed that teachers in self-contained classes offered instruction in academically oriented areas which were not included in the study. Computer lab, creative writing, and journal writing are some examples.

Discussion

Implications for practice resulting from this study include some support for self-contained instruction for 5th and 6th grade children. As hypothesized, significantly higher gains were found in three academic areas. These findings support those of Garner and Rust (1992).

There were two additional measures in the present study: transition time and instruction time. Transition time findings agree with those of Culyer (1984) and Elkind (1988) that students spend more time transitioning from class-to-class in schools which follow a departmental structure compared to those organized in self-contained groups.

One would logically assume, as Culyer (1984) did, that the self-contained structure would increase time-on-task because of the reduced time required to organize materials and change classrooms. Elkind (1988) posited, also, that the extra time spent changing classes would cut into valuable instruction time. However, the present study did not find that to be the case. The present study found no meaningful differences between departmental and self-contained situations for instruction time. Despite the longer transition time, the departmental teachers allotted a similar amount of instructional time in the five major subject areas compared to self-contained teachers. The reason for the nonsignificant difference of instruction time appeared to be that the self-contained teachers included time for computer lab, creative writing, and art. The study included some support for self-contained instruction for these children. As hypothesized, significantly higher gains were found in Language, Science and Total Battery. However, no differences were evident in Reading, Mathematics, and Social Studies.

The study was limited by the small number of classes used in one small Southern town. Observation lasted only two days for determining transition and instruction time. Future studies will need to expand this database to allow for generalization.

Table 1
Mean Value Added Scores

                   Departmentalized    Self-Contained

Test                 5th       6th       5th       6th

Total Battery        7.62     19.60     16.24     27.88
Reading             16.24     14.51     18.04     21.49
Language             2.15     12.87     20.00     18.94
Mathematics          4.98     30.90      9.49     44.47
Science             -5.60      8.16     13.28     28.22
Social Studies      23.82      5.53     21.62     19.93

References

Anderson, R. C. (1962). The case for teacher specialization in the elementary school. Elementary School Journal, 62, 253-260.

Bock, R. D., & Wolfe, R. (1996). Audit and review of the Tennessee value-added assessment system (TVAAS): Preliminary report (Technical Report, pp. 1-35). Tennessee Office of Education Accountability, Comptroller of the Treasury, Nashville, TN.

Culyer, R. C. (1984). The case for self-contained classroom. Clearing House, 57, 417-419.

Eikind, D. (1988). Rotation at an early age. Principal, 36, 11-13.

Garner, S. S., & Rust, J. O. (1992). Comparison of fifth-grade achievement in departmentalized and self-contained rural schools. Tennessee Educational Leadership, 19, 32-37.

Gibb, E. G., & Matala, D. C. (1962). Studies on the use of special teachers of science and mathematics in grades 5 and 6. School Science and Mathematics, 62, 565-585.

Lamme, L. L. (1976). Self-contained to departmentalized: How reading habits changed. Elementary School Journal, 76, 208-218.

Mac Iver, D. J., & Epstein, J. L. (1992). Middle grades education. In M. Alkin (Ed.) Encyclopedia of educational research (6th ed., pp. 834-844). New York: MacMillan/American Educational Research Association.

McWherter, N., & Smith, C. E. (1993). 21st Century Schools Value Added Assessment. Nashville: Tennessee Department of Education.

Roger, J. S., & Palardy, J. M. (1987). A survey of organizational patterns and grouping strategies used in elementary schools in the southeast. Education, 108, 113-118.

Squires, D. A., Huitt, W. G., & Segars, J. K. (1983). Effective schools and classrooms: A research-based prospective. (ASCD No. 611-83298). Alexandria, VA: Association for Supervision and Curriculum Development.

Tennessee Comprehensive Assessment Program: Guide to test interpretation. (1993). Monterey, CA: CTB, Macmillan/McGraw-Hill.

Walker, D. (1990). Fundamentals of Curriculum. San Diego, CA: Harcourt Brace Jovanovich.

Carole J. McGrath, Retired School Psychologist, Lincoln County Schools, TN. James O. Rust, Faculty Member, Middle Tennessee State University.

Correspondence concerning this article should be addressed to James O. Rust, Box 533, MTSU Station, Murfreesboro, TN 37132. Email: [e-mail].