Hei titiro anō i te whāinga: Māori achievement in bachelors degrees revisited

Introduction

This paper updates Te whai i ngā taumata atakura – supporting Māori achievement in bachelors degrees, published by the Ministry of Education with support from Te Tapuae o Rehua (Earle, 2007).

Since the regression models for the original study were run, two data improvements have been made. The matching of students in the Ministry of Education’s statistical dataset has been improved and rerun, eliminating the need for the provisional rematch used for the original models. The Ministry, with the assistance of the New Zealand Qualifications Authority, has also created a dataset that links NCEA results with tertiary enrolment data at a unit record level. There are sufficient years in this dataset to rerun the under-20-year-old first-year pass rate and return to study models using this linked data.

Also, an extra year of data is now available for each of the models. This is particularly significant for the completion models, which had previously included only one cohort of students.

This paper provides the key findings from the revised regression models. As with the original study, the models explain part of the differences in results between individuals. Adding in NCEA results improved the explanatory power of the first-year pass rates model by a small margin. Other factors, such are attitude, motivation and quality of instruction are not captured in these models and may have as much or more influence than the factors captured within the models. The models are briefly described below. A fuller description of the models and results for each effect is also available.

From school to degree study

The original models looked at Māori first-year degree students aged under 20, including those who had a gap between school and degree study. The three original models have been rerun using the revised tertiary dataset, covering:

• first-year pass rates for degree students enrolling from 2002 to 2006 inclusive (explained 15 percent of variance)
• return to study in the period from 2003 to 2006 for first-year degree students enrolled from 2002 to 2005 (explained 37 percent of variance)
• completion of a bachelors degree for students who enrolled in their first year in 2001 or 2002 and had completed at least the equivalent of three years’ full-time study¹ towards a three-year degree (explained 44 percent of variance).

These models used students’ self-reported highest school qualification. For school leavers, this information may be collected prior to the award of their final school qualification and therefore tends to understate their school achievement. It also doesn’t provide any information on how well the student did within the qualification. That is, did they just scrape through or ‘pass with flying colours’?

NCEA introduced a new set of secondary school qualifications as of 2002, replacing the previous system of qualifications. By 2004, all senior secondary students leaving school had had the chance to go through all three levels of the new NCEA system. The results for all students undertaking NCEA are collected by the New Zealand Qualifications Authority and made available to the Ministry for analytical purposes. It is possible to directly link NCEA results with tertiary education enrolments and completions through the use of the national student number.

NCEA provides an improved measure of highest school qualification of tertiary students. Using the NCEA data, rather than school qualifications collected by tertiary providers, the actual highest school qualification can be used and levels and combinations of qualifications can be more finely distinguished.

The NCEA data also provides an opportunity to look at the relative performance of students across subjects, using their results in achievement standards. A measure of achievement, called the ‘expected percentile’, has been developed for analytical purposes. The expected percentile estimates the relative performance of students in each year and for each level of NCEA by aggregating their results across achievement standards within each subject. The resulting measure is a continuous variable that estimates the performance of a student relative to their peers on a scale from 0 to 100. A fuller description of this measure can be found in Ussher (2008) and Scott (2008).

Two new models were added using NCEA results linked to tertiary results:

• First-year pass rates for degree students who left school in 2004 and 2005 and enrolled in degree study in 2005 and 2006. This includes students who left school in 2004 and had a gap year before enrolling in 2006 (explained 23 percent of variance).
• Return to study for first-year degree students enrolled in 2005 and whether they returned in 2006 (explained 34 percent of variance).

Entering degree study for the first time as an adult

The original study looked at first-time, first-year degree students who started degree study between the ages of 25 and 39. This age group was chosen to represent students who had been out of school for about 10 or more years but were still in the early to middle period of their working life.

As with the younger students, three models were rerun using the revised tertiary data:

• First-year pass rates for degree students enrolling from 2002 to 2006 inclusive (explained 22 percent of variance)
• Return to study in the period from 2003 to 2006 for first-year degree students enrolled from 2002 to 2005 (explained 34 percent of variance)
• Completion of a bachelors degree for students who enrolled in their first year in 2001 or 2002 and had completed at least the equivalent of three years’ full-time study towards a three-year degree (explained 54 percent of variance).

Footnotes 

1. The threshold used for inclusion in the model was the completion of more than 2.9 EFTS at degree level.