A new feature in National Monitoring in 1999 was the commitment to look directly at the achievement of Pacific Island students in New Zealand primary and intermediate schools. These students have been among the samples in NEMP assessments between 1995 and 1998, but not in sufficient numbers to allow their results to be reported separately. At the request of the Ministry of Education, NEMP has selected special additional samples of 120 year 4 students and 120 year 8 students to allow the achievement of Pacific Island students to be assessed and reported. The augmented sample was too small, however, to allow separate reporting on students from different Pacific Island nations (such as Samoa, Tonga, and Fiji).

All schools in the main NEMP year 8 sample that had ten percent or more Pacific Island students (as classified in school records) were selected. All other schools nationally with at least 12 year 8 students and at least 20 percent Pacific Island students in their total roll were identified, and an additional random sample of ten schools drawn from this list. A similar procedure was followed at year 4 level, except that schools already chosen at year 8 level were excluded from the sampling list. From each specially sampled school, 12 students (in three groups of four) were sampled, confirmed and assessed using exactly the same procedures as in the main sample. The students' performances were also scored in the same manner as the performances of students in the main sample.

The results for Pacific Island, Mäori, and other students in the schools with more than ten percent Pacific Island students were then compared. Because all of the schools chosen for these analyses have at least ten percent Pacific Island students, the results only apply to students at schools like these.

Differences among the three ethnic groups of students were checked for statistical significance using one way analysis of variance on the overall scores for each task.

Each analysis compared the performance of about 50 Pacific Island students, 25 Mäori students and 45 other students. The critical level for statistical significance was set at p = .05 (so that differences this large or larger among the subgroups would not be expected by chance in more than five percent of cases). Where statistical significance occurred, Tukey tests were used to identify which groups differed significantly.

The mean scores for each group on each task are presented in the tables below, together with the standard deviations for all students in this sample. Statistically significant differences are clearly indicated.

For year 4 students, there were statistically significant differences in performance among the three groups on 23 of the 50 tasks. The Pacific Island students scored significantly lower than the Mäori students on seven tasks and than the "other" students on 20 tasks. The Mäori students scored significantly lower than the "other" students on four tasks.

For year 8 students, there were statistically significant differences in performance among the three groups on 19 of the 48 tasks. The Pacific Island students did not score significantly lower than the Mäori students on any task, but were lower than the "other" students on 18 tasks. The Mäori students scored significantly lower than the "other" students on nine tasks.

Statistically significant (p<.05) differences among the means are shown thus: significantly lower or higher.