Skills and education: How well do educational qualifications measure skills?

5. People with high qualifications and low literacy and numeracy skills

This section looks at people who have degrees or higher qualifications but whose literacy and/or numeracy was measured in the ALL survey as level 1 or 2 and explores the demographic, the employment and the migration and language characteristics of this group.

Levels 1 and 2 are below what is considered a minimum for dealing with the demands of life and work in a complex, advanced society. (OECD and Statistics Canada, 2000; Statistics Canada, 2005). Level 3 is generally seen as the skill level required for successful entry to degree study. Performance above Level 2 is generally associated with a number of positive outcomes, including increased civic participation, higher earnings and more opportunities for ongoing skill development (Statistics Canada, 2002).

5.1: Demographic differences

Figure 6 below looks at the degree-holders with low literacy, broken down by age, while Figure 8 looks at differences by gender.

Low prose literacy is significantly less likely to occur among holders of degrees or higher qualifications aged over 45 years, despite there being no statistically significant difference in the proportion of people over the age of 45 with low literacy over the population as a whole (Satherley and Lawes, 2008).

While it is not possible to be definitive about the reasons for the lower prevalence of low literacy among holders of bachelors degrees aged over 45, this may be a consequence of the greater prevalence of immigration in younger age groups, meaning that the over 45 group is less likely to include people whose first language is not English. It may also reflect the fact that those in older age groups may be in more senior positions that require them to enhance and maintain their skills. Further, it may reflect the elite nature of the New Zealand university system until the 1970s. Universities tended to cater to a narrower (and hence, more able) section of school leavers until the late 1960s, when the early baby-boom generation left school. A good proportion of the degree holders in the 45-65 age group at the time of the ALL survey had been through university before the expansion of the 1970s and later.

Figure 6: Holders of a degree or higher qualification with low English prose literacy and low numeracy (level 1 or 2) by age group

Source: New Zealand results of the Adult Literacy and Life Skills Survey. Ministry of Education calculations

Note: For an explanation of the categories see Appendix A. The bars represent the margins of error (at the 95% confidence level). Margins of error are large, and more detailed breakdowns are not feasible, due to the small size of the high qualification/low skills groups.

The most obvious factor related to the age distribution of prose literacy is first language. In section 5.3 it is shown that degree holders whose first language is not English are much more likely to have low prose literacy than native English speakers. The following graph applies to all degree holders and displays the percentage in each age group who are not native speakers of English. It shows that the percentage of degree holders whose first language is not English is considerably lower in the 45-65 age group.

Figure 7: Holders of a degree or higher qualification who are not native speakers of English, by age group

Source: New Zealand results of the Adult Literacy and Life Skills Survey. Ministry of Education calculations

Note: For an explanation of the categories see Appendix A. The bars represent the margins of error (at the 95% confidence level). Margins of error are large, and more detailed breakdowns are not feasible, due to the small size of the high qualification/low skills groups.

Female graduates are slightly – but not statistically significantly – less likely to have low literacy than male, while the reverse applies in the case of numeracy.

Figure 8: Holders of a degree or higher qualification with low English prose literacy and low numeracy (level 1 or 2) by gender

Source: New Zealand results of the Adult Literacy and Life Skills Survey. Ministry of Education calculations

Note: For an explanation of the categories see Appendix A. The bars represent the margins of error (at the 95% confidence level). Margins of error are large, and more detailed breakdowns are not feasible, due to the small size of the high qualification/low skills groups.

5.2: Employment differences

Unsurprisingly, people with degrees who are in professional occupations have significantly higher literacy than those in non-professional occupations. In part, this may relate to the fact that those in professional occupations have to use (and hence, practise) their skills more. Equally, there could be a selection effect, with those with lower literacy being unable or unwilling to secure employment in a professional occupation.

Figure 9: Holders of a degree or higher qualification with low English prose literacy and low numeracy (level 1 or 2) by occupational group

Source: New Zealand results of the Adult Literacy and Life Skills Survey. Ministry of Education calculations

Note: For an explanation of the categories see Appendix A. The bars represent the margins of error (at the 95% confidence level). Margins of error are large, and more detailed breakdowns are not feasible, due to the small size of the high qualification/low skills groups.

Those with higher qualifications and lower literacy are less likely to be working for employers in community, personal and social services.

Figure 10: Holders of a degree or higher qualification with low English prose literacy and low numeracy (level 1 or 2) by industry group

Source: New Zealand results of the Adult Literacy and Life Skills Survey. Ministry of Education calculations

Note: For an explanation of the categories see Appendix A. The bars represent the margins of error (at the 95% confidence level). Margins of error are large, and more detailed breakdowns are not feasible, due to the small size of the high qualification/low skills groups.

5.3: Migration and language differences

The following graphs look at the high qualifications/low skills groups by immigrant status and by first language. Those not born in New Zealand are counted as ‘immigrants’. Immigrants constitute a diverse group – it includes refugees, skilled migrants and their families, business migrants and others, and it may include some temporary migrants. See Appendix A for further detail of immigrant categories. Different categories of immigrants can be expected to have different occupational and language skills. In this analysis, we are not able distinguish between immigrant categories and hence, the results we show below will not reflect the diversity of skills in the immigrant community.

The ALL survey was administered in English⁵, so that skills as measured in ALL are dependent on some level of fluency in English. To some extent, the New Zealand labour market discounts the qualifications of those for whom English is not a first language (Earle, 2009c; Maani, 2009), reflecting the fact that, in New Zealand’s largely mono-lingual labour market, skills in work are not necessarily revealed fully if someone is struggling with English. There may also be some reluctance in some occupations, industries or workplaces to employ people with non-native English accents. Analysis of the ALL data shows that recent immigrants with degrees are significantly more likely to have low literacy, while established immigrants with high qualifications are significantly more likely to be in this category than people born in New Zealand.

The relationships between literacy and numeracy skills and immigrant status are not necessarily direct. Due to the small size of the high qualifications/low skills groups, there is little scope for more in-depth analysis of different immigrant categories within these groups. However, one of the main factors clearly related to English literacy and numeracy is first language, and one of the main differences between the different immigrant statuses is the proportions of degree-holders with English as a first language: for the New Zealand born, this proportion is 98 percent; for established immigrants, it is 52 percent; and for recent immigrants it is 29 percent.

Figure 11: Holders of a degree or higher qualification with low English prose literacy and low numeracy (level 1 or 2) by migration status

Source: New Zealand results of the Adult Literacy and Life Skills Survey. Ministry of Education calculations

Note: Bars represent margins of error (at the 95% confidence level). Margins of error are large, and more detailed breakdowns are not feasible, due to the small size of the high qualification/low skills groups.

As implied in Figures 6 and 7 above, first language is an even more powerful predictor than immigrant status of low English prose literacy among holders of degrees or higher qualifications. About two-thirds of degree holders with low English prose literacy have a first language other than English; 33 percent have English as a first language, 45 percent have an Asian first language and 22 percent another first language. About half of degree holders with low numeracy have a first language other than English; 47 percent have English as a first language, 33 percent have an Asian first language, and 20 percent another first language. Figure 12 shows that those with English as a first language who hold a degree or higher qualification are relatively unlikely to have low English prose literacy.

Figure 12: Holders of a degree or higher qualification with low English prose literacy and low numeracy (level 1 or 2) by first language

Source: New Zealand results of the Adult Literacy and Life Skills Survey. Ministry of Education calculations

Note: For an explanation of the categories see Appendix A. The bars represent the margins of error (at the 95% confidence level). Margins of error are large, and more detailed breakdowns are not feasible, due to the small size of the high qualification/low skills groups.

These two findings reflect two features of recent immigration trends. The first is the fact that many immigrants have higher qualifications. Immigrants are roughly twice as likely to have a degree as the New Zealand born (Maani, 2009), reflecting preference given to immigration applications from those with degrees over much of the last 15 years. The second is that many immigrants come from countries where English is not the first language⁶.

Footnotes

5. Literacy and numeracy were measured in ALL using a standardised international instrument, rather than one tailored specifically to New Zealand social and cultural requirements. However, analyses of the relation between literacy or numeracy as measured in ALL and labour market outcomes (Satherley, Lawes and Sok 2008b, Earle 2009b) indicate that the skills measured by ALL approximate the English literacy and numeracy skills demanded in the New Zealand labour market.
6. This paper does not traverse the changes in immigration policy over the years. Changes made to immigration policy since 2004 have given greater weight to matching skills of applicants to skill needs in the economy. This may tend to have different effects over time.