Timeframe
1995–
COMPASS staff
Peter Davis
Barry Milne
Natalia Boven
Nichola Shackleton
COMPASS students
Bryan Byun
Katie Fahy
Emily Ren
Collaborators
University of Auckland
Alan Lee
Andrew Sporle
Statistics New Zealand
Pat Coope
Description
The New Zealand Socioeconomic Index (NZSEI) was inspired by the International Socioeconomic Index described in 1992 (Ganzeboom HBG, De Graaf PM, Treiman DJ 1992). It uses a “returns to human capital” model, which posits that there is a relationship between cultural capital and material rewards, and that this relationship is mediated by occupation. More simply, it views occupation as the means by which one’s education is converted into income, and so sees differences in occupation as likely to represent differences in life chances and opportunities. On that basis, the model uses occupation to create a set of socioeconomic strata.
The forerunner of the NZSEI was the widely-used Elley-Irving scale, which assigned every occupation to one of six socioeconomic status groups based on equal weightings of education level and income associated with it. The NZSEI is grounded in a slightly different conceptual framework, and was an attempt to derive for New Zealand an occupation-based measure of socioeconomic that could be used as both a continuous variable and a grouping variable. The original derivation, and iterations through 2001 worked with the New Zealand Standard Classification of Occupations (1988, 1995, 1999); more recent updates have used the Australian and New Zealand Standard Classification of Occupations.
The NZSEI was first developed using 1991 Census data (Davis, et al. 1997) and it has subsequently been updated for the 1996 Census, the 2006 Census, the 2013 Census, and the 2018 Census. The 2006 and 2013 versions were developed largely by summer scholars working with COMPASS, and the update for the 2023 Census is likewise being finalised following the work of Emily Ren, one of our 2024-2025 summer scholars.
In terms of its construction the NZSEI is a very convenient measure, requiring only occupation data, so it can be readily applied to many administrative datasets and social surveys. Its continuous scale runs from 10 (low socioeconomic status) to 90 (high socioeconomic status), and scales constructed for the 1996 Census and beyond also include a method for imputing occupational scores for people not in the labour force, based on age and education.
All versions of the NZSEI have been shown to validate well against several health and social outcomes. NZSEI-06 and NZSEI-13 showed expected socioeconomic gradients with area deprivation, smoking prevalence, and housing tenure. NZSEI-06 was also tested against access to a motor vehicle and showed a slight relationship therewith. NZSEI-18 was further validated against hospitalisations, and self-rated health and life satisfaction, and showed the expected socioeconomic gradients, of varying strengths.