SCHOOL SIZE AND STUDENT ACHIEVEMENT
Terry Crooks and Katherine
Hamilton
Because
much of New Zealand is sparsely populated, it has a very high
proportion of small schools. The Ministry of Education and the
Education Review Office have suggested that very small schools
face major challenges and often perform less well than their larger
counterparts. These concerns have not been based on student achievement
data. This study compared the achievement of students from schools
of different size, using data from all NEMP assessment tasks (covering
15 curriculum areas) administered to Year 4 and Year 8 students
between 1995 and 2000. |
| One-way
analysis of variance was used to compare the students’ task
performance across the different school sizes. At Year 4, the NEMP
definitions of school size were ‘small’ (4 to 19 Year
4 students), ‘medium’ (20 to 35), and ‘large’
(more than 35). There were 144 small schools, with a median total
roll of 74 students. Twenty-six percent of these schools had fewer
than 50 students. At Year 8, the NEMP definitions were ‘small’
(4 to 34 Year 8 students), ‘medium’ (35 to 150), and ‘large’
(more than 150). There were 209 small schools, with a median total
roll of 138 students. Fifteen percent of these schools had fewer than
50 students. |
A second stage of analysis used only students in the small schools, now subdivided into ‘very small’ and ‘quite small’. At Year 4, very small schools had 4 to 12 Year 4 students and quite small schools had 13 to 19 Year 4 students. The median total roll for these schools was 51 students, with 25 percent of the schools having fewer than 40 students. At Year 8, very small schools had 4 to 16 Year 8 students and quite small schools had 17 to 34 such students. The median size for very small schools was 92 students, with 27 percent of these schools having fewer than 60 students. |
•
Comparisons across the small, medium and large schools yielded significant
performance differences on only 2% of the tasks at both year levels.
• At Year 4, students in very small schools performed better than
students in quite small schools on 3% of the tasks, equally well on 91%
and less well on 6%. On average, across all tasks, a median student in
the very small schools performed as well as or better than 48% of the
students attending quite small schools.
• At Year 8, students in very small schools performed better than
students in quite small schools on 6% of the tasks, equally well on 91%
and less well on 3%. On average, a median student in the very small schools
performed as well as or better than 51% of the students attending quite
small schools.
| The
results provide strong evidence that, within the range of school sizes
explored, school size is not an important factor influencing student
achievement at Years 4 and 8 in New Zealand schools. However, relatively
few of the ‘very small’ schools were sole-charge schools,
and a substantial proportion had three or more classes, so there is
still a possibility that |
extremely small schools could produce weaker student achievement outcomes than larger schools. While the available research provides some legitimate arguments for closing or amalgamating very small schools, poor student achievement in such schools is not one of them. |