Social and economic indicators of education

2. Introduction

Purpose

The aim of this report is to present new information from the first New Zealand General Social Survey (NZGSS) on the relationship between education and a range of social and economic indicators.

The results in this report are based some 6,400 adults in this survey representing the population of nearly 2.4 million New Zealanders aged between 25 and 64. This report explores 30 social and economic outcomes as they relate to one's highest level of educational attainment.

The analysis is descriptive in nature, and does not, in general, explore underlying factors, causal directions or the existing literature in depth. Instead, it aims to tap into the rich new information collected in New Zealand's first General Social Survey to provide a broad picture of how education is associated with a range of social and economic indicators.

While it is accepted that people without a secondary education qualification are disadvantaged economically, this report looks at the impact a lack of secondary qualifications has on other aspects of life. It also explores what differences there might be between those who leave with just one-year upper secondary qualifications, and those who have higher-level school qualifications. Much of the Government's recent strategic focus in tertiary education has been on getting more people studying at level 4 and above. This has included reviews of the relevance and funding of level 1 to 3 post-secondary provision. This report provides further information to support the benefits or otherwise of level 1 to 3 qualifications. There are significant known financial benefits for those gaining bachelors or postgraduate qualifications. To what extent does this extend to other non-financial areas, and to what extent do these benefits remain after you adjust for higher employment and earnings?

Background

There is an established literature showing the benefits of education on employment and earnings. However, less knowledge exists on the non-financial or social benefits of education. While policy interest has traditionally focused on these financial and economic outcomes, interest in the contribution education makes to improved social outcomes is growing.

This interest is growing as new international survey collections are expanding the pool of data now available. An indicator on social outcomes of education was included for the first time in 2009 in the annual OECD indicators publication, Education at a Glance (OECD, 2009). This compared the gains in health status, interpersonal trust and political engagement across education levels for OECD countries. The European Social Survey was first introduced in 2002 and, now into its fifth round, covers over 30 countries. The International Social Survey Programme and World Values Survey also provide internationally established collections of social data, where indicators can be related to a range of background information, such as education, and viewed in a domestic or international context.

An earlier Treasury report (Johnston, 2004) undertook a review of the wider benefits of education. This report reviewed the evidence that greater education causes better outcomes in life, over and above the effects of having a higher-paying job. The report says: "Research suggests that increased education, as measured by the time people spend in formal education or the qualifications they attain, may cause a reduction in cigarette smoking, anxiety disorders, anti-social disorders, suicide, crime, teenage pregnancies, unemployment and reliance on welfare benefits, at least when these outcomes are measured in young adulthood. Education may also have an effect on people’s health. The wider benefits of education are difficult to quantify, however, and the degree of uncertainty around them is considerable. Policy-makers would be unwise to rely too heavily on the existence of wider benefits when making decisions about public investment in education."

The Ministry of Education has recently used NZGSS data to examine the benefits of sub-degree study for 25 to 39 year-old adults (Earle, 2010). This report showed, among other things, that level 1 to 3 post-school certificates were associated with lower employment and income than school qualifications, but were better than having no qualifications at all, while the social outcomes for people with level 1 to 3 certificates were similar to those of people with no qualifications. Women with diplomas were likely to have better health, higher overall life satisfaction, and were more likely to volunteer and to read to their preschool children than women with no qualifications. These effects were less evident for men. The report goes on to say: "There is evidence that education can have an effect on social outcomes over and above, or independent of, its effect on improving employment and income."

An earlier Ministry of Education report also collated evidence around outcomes of the New Zealand tertiary education system (Smart, 2006). The report included information on employment and income, economic growth, health, living standards and crime. It concluded that: "Overall, the overwhelming weight of evidence points to there being significant and positive outcomes from the tertiary education system."

While the results from these studies, and this report, provide additional support for the benefits of education, beyond employment and earnings, care is needed when making inferences from these results and deciding the extent to which they can inform policy. The direct contribution of education to these benefits is by no means established. It is not clear what degree of causal link exists between education and improved social outcomes, and how much is mediated through factors such as employment and income, age, gender, income and immigrant status.

While education may be a factor leading to improved outcomes in some areas, in other areas it may be the poorer outcome that is acting as a barrier to further education. For example, while higher levels of education are associated with better health, it is less clear how much education plays a role in this, either directly or indirectly or, in fact, how much better health aids improved education. In some cases, sample errors prevent conclusive inferences being drawn about differences.

But regardless of the contribution or otherwise of education to these improved outcomes, the increasing evidence base on these wider social and personal outcomes does provide at least suggestive evidence for policy makers, students and parents looking to shape education investment decisions. Apart from the intrinsic social benefits that may be sought, there may also be indirect financial benefits accrued, over and above a better-paying job. For example, the public and private health cost savings associated with reduced smoking, or reduced incidence of depression. Prospective students' decisions are often based on financial considerations, and even where there may be non-financial expectations (such as better health and more leisure time), these are also often expected to be largely realised via increased earnings.

The New Zealand General Social Survey

This report uses data from the New Zealand General Social Survey (NZGSS). This survey was run by Statistics New Zealand for the first time over the period April 2008 to March 2009. It surveyed over 8,700 people, collecting answers to around 140 social- and economic-related questions, along with more than 50 background demographic questions about the person and the household. The survey provides a valuable new resource for examining the relationship between education and a range of social and economic outcomes never before collected together in such a way.

This report is limited to those adults aged 25 to 64. This covers just over 6,400 survey respondents, representing an estimated population of nearly 2.4 million adults. Excluding under 25-year-olds recognises that many of this group are still undertaking education, and often the benefits of any education they have attained haven't yet been fully realised. Adults aged 65 and over are also excluded, as it is difficult to realistically compare outcomes for this group where leaving school early – often without qualifications – was the norm, and for whom age is likely to be more of a factor in many of these indicators (eg health and labour market outcomes).

The first release of NZGSS data was in October 2009 (Statistics New Zealand, 2009). This release included some information by education, showing positive associations between education and income, health, voluntary work and household recycling, and a slightly higher level of satisfaction with life.

This release said "…The NZGSS is an important monitoring tool for areas that receive considerable government funding. Increasingly, government seeks to act in a coordinated way across agencies and this survey contributes to that goal. In particular, the survey enables a view of those with multiple good and/or bad outcomes."

The indicators

This report includes 30 of the questions available from the NZGSS. These questions have been grouped under 18 indicator headings in this report. For the purposes of this report, these questions were selected under a framework of five broad areas: economic (employment, income, economic standard of living), social (marital status, social contact, tolerance, voting, volunteering, national identity), health and safety, environmental practices, and overall life satisfaction. There are many other questions that could have been included, and it is recognised that this report does not cover the rich spectrum of topics available from the NZGSS. For example, the report does not look at housing or leisure, or a range of questions on knowledge and skills. However, it is hoped that the selected indicators do provide useful coverage of many of the social and economic indicators that may be of interest.

Education levels

This report explores the relationship between 30 social and economic indicators across eight types or levels of education. Each level relates to a person's highest level of educational attainment. The highest level has been derived from two separate questions in NZGSS on highest school qualification and highest post-school qualification. Aggregating different types and levels of qualification needs to balance the risks of loss of information when different groups are combined, with the need to have a large enough sample size, and to be able to present a simple descriptive story. In line with these considerations, and the standards used for other Ministry of Education analyses, the following groupings were chosen:

1. No qualification
2. Level 1 upper secondary qualification (eg NCEA 1, school certificate)
3. Level 2 to 3 upper secondary qualification (eg NCEA 2-3, University Entrance)
4. Level 1 to 3 post-school certificate
5. Level 4 certificate
6. Diploma
7. Bachelors
8. Postgraduate

The levels refer to levels in the New Zealand Register of Quality Assured Qualifications – which range from 1 (generally equivalent to year 11) to 10 doctorate (see http://www.kiwiquals.govt.nz/about/levels/index.html). Figure 2 shows the distribution of adults in the NZGSS across these highest education attainment levels. 

Figure 2: Percentage of 25 to 64-year-olds in the NZGSS by highest level of education attainment

Error bars show 95% confidence intervals.

The extent to which staying an extra year or two in upper secondary makes a difference is explored explicitly in this report via the level 1 and level 2-3 upper-secondary level split. Generally, at least two years of upper secondary education is required to gain access to higher levels of education. Compared with other OECD countries, New Zealand has a much higher proportion of people who leave after one year of upper secondary (OECD, 2009, Table A2.1).

Around 25% of adults aged 25 to 64 had an upper secondary school qualification as their highest qualification. This comprised around 12% with a level 1 equivalent school qualification and 13% with a level 2-3 equivalent school qualification. These two groups included around 4% of adults who had an overseas-based school qualification as their highest level of attainment. Results for the group with overseas school qualifications as their highest attainment were noticeably poorer than for those with NZ school qualifications. The inclusion of overseas-based school qualifications with the NZ school results often reduced indicator results by around 2 or 3 percentage points. This is likely to reflect factors other than educational ones, but this is not explored in this report.

It is not possible from the NZGSS to distinguish how many of the overseas school group have a school qualification equivalent to level 1 and how many have a school qualification equivalent to level 2 or 3. However, we still wish to include this group in such a way so as to retain the opportunity to explore differences between level 1 and level 2 or 3 school qualifications, and to ensure that school and tertiary results are provided on a comparable basis.  

To enable this, we treated the overseas school group in the following way. We first assumed that all overseas school qualifications were equivalent to level 2 or 3 school qualifications. We then assumed that none were equivalent to level 2 or 3 school qualification. We took these two scenarios as a reasonable approximation of the worst and best case scenarios for the true split of the overseas school group between the two school levels. We then used the mid point of these extremes as our estimate for this report. The difference between this midpoint and the best or worst case results was added to the existing sample confidence intervals to provide an extended estimate of the uncertainty around the true value for these two education levels.

While level 1 to 3 post-school certificates are at the same educational level as school level 1 to 3 qualifications, the outcomes for this group are quite different. This is shown, for example, in Earle (2010), and is reflected again in this report. While level 1 to 3 certificates can have a vocational focus (especially at level 3), they also often have a foundation education focus (especially at levels 1 and 2) – with content related to literacy, numeracy or generic life skills. While outcomes for this group were better than for those with no qualifications, they were often worse than those for whom level 1 to 3 school qualifications are the highest level of attainment.

However, this may reflect a selection effect, rather than an education effect. Those who undertake level 1 to 3 post-secondary study are mostly those who did poorly at school – that is they have no school qualifications. There may well be factors other than education that are leading to these poorer outcomes. We don't really know what the outcomes would have been if this group had not obtained their certificate. Similarly, those that undertook level 1 to 3 post-secondary study and then successfully moved on to higher-level study are not included in this group. Results may then, in fact, reflect a kind of double-selection effect as described, for example, in Earle (2010).

While the majority of those with level 1 to 3 post-school qualifications had no school qualifications, there were some who did have school qualifications at level 1 to 3 as well. When we analysed outcomes for this group, results were often more in line with the post-school group than the school group. That is, their results were often lower than the school groups but higher than the no qualifications group. For this reason, those with both school and post-school qualifications at level 1 to 3 were included in the post-school level 1 to 3 group for this study.

About 136,000, or 6%, of the adults aged 25 to 64 reported 'other' post-school qualifications. An examination of the profile of, and outcomes for, these adults showed a similarity with the post-school level 1 to 3 certificate group. The NZGSS question provides tick box categories for National Certificates only, so this may have led respondents with non-national level 1 to 3 certificates to use the 'other' category.  So for this report, a decision was made to incorporate all those reporting 'other post-school qualifications' into the level 1 to 3 group. While this may be incorrect for some of these 'other' qualifications, some analysis of the resulting outcomes for the combined 'level 1 to 3 post-school/ other' group suggests a more predictable fit.

Figures 3 to 5 show the education attainment distribution of 25 to 64 year olds by sex, age and immigrant status. While men were more common in the vocational group, women were slightly more common in the level 1 to 3 post-school certificate group. At other levels there was not much difference. Younger adults were noticeably more likely to be higher-qualified than older adults, reflecting a shift in educational expectations over the years. Overseas-born adults were more likely to hold degrees than New Zealand-born, reflecting immigration policy. This was significant enough to distort some indicator results, in particular those relating to tolerance and contact with family and friends.

Figure 3: Percentage of 25 to 64-year-olds in NZGSS by education level and sex

Figures are row percentages. Each row adds to 100%. Error bars show 95% confidence intervals.

Figure 4: Percentage of 25 to 64-year-olds in NZGSS by education level and age group

Figures are row percentage. Each row adds to 100%. Error bars show 95% confidence intervals.

Figure 5: Percentage of 25 to 64-year-olds in NZGSS by education level and whether NZ-born or not

Figures are row percentages. Each row adds to 100%. Error bars show 95% confidence intervals.