Wisconsin Study Finds 'Time on Task' Has Small Effect on Pupil Achievement
Madison, Wis--Increasing students' "time on task" is a desirable goal, but it may not be the panacea some educators say it is, according to a University of Wisconsin researcher. In fact, Richard Rossmiller, chairman of the university's department of educational administration, has found a relatively weak link between time on task and student achievement in reading and mathematics.
Speaking this month at the Wisconsin State Superintendent's Conference for District Administrators, Mr. Rossmiller suggested that time on task is "probably a better descriptor of a student than a predictor of the student's academic achievement."
Mr. Rossmiller drew his conclusions from an analysis of the data he obtained in a 3-year study of a representative sample of Wisconsin elementary-school students. Students in four schools--two in medium-sized urban areas and two in small districts--were observed through their 3rd-, 4th-, and 5th-grade years.
For three days each year, each student was observed, and his activity recorded, at three-minute intervals. Observers noted the academic subject, the mode of instruction, and whether the student was "on task,'' "off task," or engaged in "process behavior"--such as waiting for the teacher to give instructions or grade papers. Academic progress was measured annually by the students' scores on the Stanford Achievement Test.
The most disappointing finding, the researcher said, was the amount of time actually spent on task, regardless of the mode of instruction.
The 3rd graders spent 74 percent of their time engaged in instruction, 4th graders 77 percent, and 5th graders 78 percent. Too much time, he said, was wasted in preparation and waiting.
"The amount of time lost to process behavior suggests this is an area where principals and supervisors can provide real assistance by helping teachers improve classroom skills," said Mr. Rossmiller.
The amount of class time spent on reading, mathematics, language arts, science, and social studies decreased between the 3rd and 5th grades.
Third graders spent an average of 210 minutes per day on those subjects, while 5th graders spent only 154 minutes daily.
Most on-task time was spent in independent study or large-group instruction; very little time was devoted to one-on-one instruction with the teacher or working with other students.
Mr. Rossmiller then examined the relationship between reading scores and the way students spent their time. He found that time on task accounted for 20 percent of the difference in reading scores in 3rd graders, 25 percent in 4th graders, and 16 percent in 5th graders.
These findings are significant, Mr. Rossmiller said, "but it's evident that many factors other than time on task account for the variance in the academic achievement of students."
The type of on-task time observed--independent study, one-to-one instruction, small- and large-group instruction or student-to-student instruction--may describe the student, but it does not predict academic achievement, Mr. Rossmiller said.
The most able students probably spend more time in independent study, he said, because they complete their assignments faster than other students. Such students are also more likely to work with other students in a tutorial capacity. Less able students, he suggested, tend to be involved more often in one-to-one and small-group instruction because the teacher may feel they need additional help.
Time on task was a more useful predictor of reading scores for students in the bottom quartile than for those in middle half and upper quartile. On-task time proved to be a poorer predictor of mathematics scores than of reading scores for lower-quartile students.
These results, while preliminary, indicate that "time on task is not a cure-all for the educational problems of all students," Mr. Rossmiller concluded.
Past research, he added, has indicated that classroom management, along with student-body composition, academic emphasis, and discipline, has had a greater bearing on students' achievement than have physical facilities, school and class size, organizational structure, and level of spending per pupil.
Further analysis will include data collected from parents and guardians on students' background and personal characteristics.
Vol. 02, Issue 04