An examination of the links between parental educational qualifications, family structure and family wellbeing 1981-2006

Publication Details

The primary purpose of this report is to examine and describe the relationship between family structure and family wellbeing and the educational qualifications of parents in New Zealand families over the period 1981–2006.

Author(s): Gerard Cotterell, Martin von Randow and Mark Wheldon, The University of Auckland.

Date Published: September 2008

Please consider the environment before printing the contents of this report.

This report is available as a download (please refer to the 'Downloads' inset box).  To view the individual chapters please refer to the 'Sections' inset box.

Appendix C: Wellbeing indicators

Focusing on parents

The indicators used in this analysis are based on whether at least one of the parents (in a couple) or the parent (in a single-parent family) has a particular attribute. This was based on the assumption that the income, educational and employment characteristics of parents were likely to be of greater relevance to assessing family wellbeing than those of children.

Indicator rationalisation

The wellbeing indicators recommended by Milligan et al. were constructed (see (Milligan et al. 2006), Table 10.1), but preliminary analyses suggested that certain modifications should be made to aid overall understanding and to increase indicator relevance. These modifications are discussed in the subsections below. In particular, having two sets of indicators − one at the family level and one at the household level − provided many overlapping measures and made interpretation cumbersome. These were reduced to a single set by re-defining the household-level indicators to apply to family units. In order to provide extra contextual information and comparison groups, indicators for the one person and multi-person household composition categories were calculated separately and included in the analyses.

The original set of indicators (Milligan et al. 2006) contained some in which a high score indicated a negative impact on wellbeing (e.g., hours worked) and some in which the opposite was true (e.g., post-secondary educational attainment). This has now been rationalised to improve indicator interpretation: a 'high' score on all proportion-based indicators signifies a negative impact on wellbeing. The exceptions to this rule are the equivalised income and income inequality indicators, which are not measured by proportions.

Individual- to family-level variables

The proportion-based indicators recommended by (Milligan et al. 2006) were defined using a simple 'one is sufficient' principle, whereby a family/household was counted in the numerator if at least one member possessed the relevant characteristic. For example, in calculating the post-secondary educational attainment indicator, all families where at least one member possessed a post-secondary qualification were included in the numerator.

Retirement status

During initial investigations of indicator results, a clear distinction arose between couples and single people of retirement age and those who were younger. Therefore, retirement status was added to the family/household classification scheme. The retirement status classification is defined by age and receipt of New Zealand Superannuation. It is only defined for couple-only families and one-person households. The correspondence between the retirement status categories and census variables is shown in Appendix Table C.1. A couple-only family was deemed 'retired' if both partners were aged 65 years or over and/or both partners received New Zealand Superannuation, or if one partner was aged 65 years or over and the other received New Zealand Superannuation. Similarly, one-person households were deemed 'retired' if the householder was aged 65 years or over, or received New Zealand Superannuation.

Compulsory retirement ages were made unlawful in New Zealand in 1999 (Statistics New Zealand, 2004a) and, to our knowledge, no official statistical definition of retirement exists. The definition used here was chosen as a simple and easily applied operational classification; it is not intended to be definitive. Indeed, overseas researchers have noted that a robust definition of retirement is difficult to specify (for example see Bowlby 2006). Statistics Canada has an official definition that sets 55 as the minimum retirement age and includes labour-force and income source characteristics to refine the classification (Bowlby 2007). In New Zealand, as in Canada, however, the age at which people cease work or change their working patterns varies among people and over time (Gower 1997; Statistics New Zealand 2004).

Appendix Table C.1: Retirement status classification, by family/household type, age and receipt of NZ Superannuation
Family/Household Type Age and Receipt of New Zealand Superannuation
Retired Couple only (family) Both partners are aged 65 years or over and/or receive New Zealand Superannuation
One-person household Aged 65 years or over or receives New Zealand Superannuation
Non-retired Couple only (family) Only one partner is aged 65 years or over and/or in receipt of New Zealand Superannuation, or neither is
One-person household Aged less than 65 years and does not receive New Zealand Superannuation

Added indicators

Families without paid work

The unemployment and income source indicators originally proposed by (Milligan et al. 2006) were based on the premise that unemployment and benefit receipt may be correlated with reduced wellbeing. However, changes in the benefit system between 1981 and 2001 made it difficult to construct a comparable time series for income sources, even after certain transfers such as the Family Benefit and New Zealand Superannuation were omitted. In response to this, the 'families without paid work' indicator was created. It replaces the unemployment and income source indicators.

Contact Us

Education Data Requests
If you have any questions about education data then please contact us at:
Email:      Requests EDK
Phone:    +64 4 463 8065