Executive Summary

Key findings

Factors most strongly associated with document literacy skills for people aged 16-24

Factors most strongly associated with document literacy skills for people aged 25-65

Contribution of this report

The Adult Literacy and Life Skills (ALL) Survey 2006 measured the literacy and numeracy skills of New Zealanders aged 16-65. A series of previous studies of the survey data have shown differences in literacy and numeracy according to a range of individual characteristics, including completed education, age, gender, geographical region, ethnicity, immigrant status, first language, labour force status, occupation and industry.
This report considers these factors along with a number of additional factors, and seeks to develop an overall view of how strongly these factors relate to document literacy and how the factors interact with each other, for people aged 16 to 24, in comparison to people aged 25-65.

This report adds to previous work in three ways:

• by highlighting the factors associated with the literacy skills of people aged 16-24
• by analysing the effects of a number of previously under-studied variables, namely:
  • purposes of computer use
  • mobile phone use
  • watching television and videos
  • library use
  • number of books in the home
  • patterns of personal reading
• by constructing comprehensive statistical models (based on ordinary least squares regression) in order to identify which variables are most directly associated with document literacy for the 16-24 and 25-65 age groups

This report complements the report Literacy skills of young adult New Zealanders, which is also based on the ALL survey.

Models for document literacy

This report focuses on document literacy, which is the ability to interpret non-linear texts such as tables and diagrams.

Statistical models for document literacy scores in the 25-65 and 16-24 age groups were developed including the following potential explanatory factors:

• demographic and home background factors:
  • age
  • gender
  • ethnic identification
  • first language and main language spoken at home
  • socioeconomic deprivation
  • parents' education
• education factors:
  • level of education completed
  • recent formal or non-formal learning
  • a measure of school achievement
  • experience of New Zealand education
• employment-related factors:
  • labour force status
  • occupation
  • industry
• technology use and literacy-related activities:
  • home and work computer use
  • computer use for writing or editing
  • mobile phone use
  • watching television and videos
  • library use
  • number of books in the home
  • patterns of personal reading

For the 25-65 age group, previous work has identified three key factors related to literacy and numeracy:

• language (especially first language learned at home)
• completed education
• computer use (especially at work)

This report identifies three additional factors which are also strongly associated with document literacy for the 25-65 group (when the effects of language, education, computer use and other variables are already accounted for):

• the number of books in the home
• particular combinations of language and ethnic identification (Māori and Pasifika who speak English most often at home, compared with Europeans who speak English most often at home)
• particular combinations of age and gender (older women compared with younger men)

For the 16-24 age group, the factors most strongly associated with document literacy (in a comprehensive statistical model including all the factors listed above) are:

• language spoken most often at home
• Māori or Pasifika ethnic identification (where English is the language spoken most often at home)
• educational participation and achievement
• home computer use
• library use

Building the statistical models

In order to gain some understanding of the connections between different potentially explanatory factors, separate models were constructed for the relationships between document literacy scores and four sets of factors:

1. Demographic and home background factors
2. Education factors
3. Employment-related factors, and
4. Technology use and literacy-related activities

The comprehensive models were constructed by combining these models sequentially in the order 1, 2, 3, 4.

For both the 16-24 and 25-65 age groups, within each of the four separate models, almost all the factors had significant associations with document literacy.

The exceptions to this generalisation within the 25-65 age group were: gender within the 25-44 age subgroup, current student status, and television/video watching.

There were more exceptions within the 16-24 age group, namely: age, gender, first language, mother's qualifications, non-degree tertiary study (compared with upper secondary), recent non-formal learning, experience of New Zealand education, frequency of using a computer for writing or editing, mobile phone use, and patterns of personal reading. One reason there were fewer significant associations in this age group was that this age group is not as heterogeneous as the 25-65 age group. It is clearly not as heterogeneous in age. Similarly, almost all those aged 16-24 had experience of New Zealand education, and almost all were mobile phone users, while few had been involved in non-formal learning, which means that people in this age group are not differentiated by these factors. Another reason could be that the size of the subsample for this age group is smaller and so factors with subtle effects may not be identifiable as significant.
When the four separate models are combined for each age group, a number of the factors no longer appear to be significant, because their effects are now better accounted for by other factors or combinations of factors.

For example, in the models based on demographic and home background factors, parents' education has strong associations with document literacy scores; but in the combined models, this factor is no longer significant. This appears to be because parents' education approximately predicts respondents' educational outcomes, while it is the respondents' education that is more directly associated with their literacy scores. Similarly, for the 16-24 age group, the effect of the socioeconomic deprivation of the neighbourhood is subsumed in the effects of respondent education variables.

Because education is an important factor affecting respondents' labour force status, their occupation, and the industry they have been employed in, some of the education and employment-related variables become non-significant in the combined models. For the 25-65 age group, non-degree tertiary education (compared with upper secondary) and recent non-formal learning are no longer significant; nor are labour force status and industry. For the 16-24 age group, having Year 11 as the highest completed education level is no longer significant; nor are labour force status, occupation or industry. The loss of significance of employment-related variables in the 16-24 age group is not too surprising, considering that many people in this age group are only marginally involved in the labour market, and if they have employment it may not reflect their literacy skills.

For the 25-65 age group, technology use and literacy-related activities (apart from television/video watching) remain significantly associated with document literacy in the full combined model. However, for the 16-24 age group, the only variables in this set which are significant in the full combined model are home computer use and library use. Computer use at work, computer use for writing or editing, and books in the home are no longer significant: their effects must be accounted for by home computer use and library use, and/or by the effects of home language, ethnic identification and education.

Key factors for people aged 16-24

For the 16-24 age group, the full combined model accounts for 39 per cent of the variation in document literacy scores in this age group. A reduced model based on

• main home language
• main home language/ethnic identification
• formal education
• home computer use

accounts for 32 per cent of the variation.

For people aged 16-24, the factors in the reduced model are related to document literacy in the following ways:

• Document literacy scores are negatively associated with speaking a language other than English most often at home (compared with speaking English most often at home).
• Among people who speak English most often at home, having Māori or Pasifika ethnic identification is negatively associated with document literacy (compared with not having Māori or Pasifika ethnic identification).
• Highest completed education of Year 10 or less is negatively associated with document literacy, while having completed a degree or recently studied at degree level is positively associated with document literacy (compared with having highest completed education of upper secondary).
• Document literacy scores are positively associated with achieving (self-reported) good grades in mathematics at school. Mathematics grades are likely to be indicative of overall school achievement.
• Document literacy is positively associated with using a computer at home for 5 or more hours per month (compared with using a home computer for fewer hours or not having access to a computer at home). This may well be because computer use provides practice which enhances document literacy.

Key factors for people aged 25-65

The full combined model for the 25-65 age group accounts for 48 per cent of the variation in document literacy scores in this age group. A reduced model based on the factors with the strongest effects, covering

• first and main home language
• home language/ethnic identification
• age/gender
• formal education
• work and home computer use
• books in the home

accounts for 45 per cent of the variation.

For people aged 25-65, the factors in the reduced model are related to document literacy in the following ways:

• Document literacy scores are negatively associated with having a language other than English as a first language, or as the language spoken most often at home (compared with having English as a first or main home language).
• Among people who speak English most often at home, having Māori or Pasifika ethnic identification is negatively associated with document literacy (compared with not having Māori or Pasifika ethnic identification).
• Document literacy scores are negatively associated with being female, especially female aged 45-65, compared with being male aged 25-44.
• Highest completed education of Year 11 or less is negatively associated with document literacy, while having completed a degree or having studied recently at degree level is positively associated with document literacy (compared with having highest completed education of upper secondary).
• Document literacy scores are positively associated with using a home computer at least 5 hours per month (compared with using a home computer for fewer hours or not having home computer access). Document literacy is also positively associated with having used a computer at work in the past year (compared with being employed but not using a computer at work, or not having been employed).
• Having 25 or more books in the home is positively associated with document literacy scores (compared with having fewer than 25 books).