Promoting quality research
This report used data from the Performance-Based Research Fund (PBRF) Quality Evaluations to analyse the factors associated with the likelihood of university academics being promoted between 2003 and 2006.
Author: Warren Smart, Tertiary Sector Performance Analysis & Reporting, Ministry of EducationDate Published: April 2009
2 - Introduction
In a recent report, Gibson, Tressler and Anderson (2008) look at the factors linked to career advancement among New Zealand university economists. They claim that there appears to be a lower degree of association between the academic rank of university economists and the quality of their research following the introduction of the Performance-Based Research Fund (PBRF). Instead, their research shows a stronger association between quantity of research and academic rank following the introduction of the PBRF. Gibson et al. suggest that if this is also the case in the other academic disciplines, then this would imply that there are low private benefits from improving research quality. Over time, that could undermine the main objective of the PBRF – which was to raise the average quality of research.
The British researcher Jonathan Adams, who conducted an independent review of the PBRF for the Tertiary Education Commission and the Ministry of Education in 2008, also stressed the importance of the individual benefits from improving research quality. He stated that:
“To support and comply with the underlying policy, people need to believe that their personal investment will lead to beneficial changes to their research environment, at whatever level of granularity.” Adams (2008: 37).This report uses data from the PBRF Quality Evaluations to analyse if there is a relationship between higher research quality and the likelihood of being promoted to a higher academic rank in a university. The analysis applies logistic regression to the PBRF unit record data to isolate the effects on promotion of research quality from the effects of other factors, such as age.
The structure of the report is as follows. In section 3, background on the introduction of the PBRF is presented. Then, in section 4 the dataset used in the analysis is discussed and the logistic regression model presented. The results of the logistic regression are then presented in section 5. Finally, some conclusions are presented in section 6.


