CAL logo

'Surveys and Beyond: The case for adult literacy'


National surveys and national campaigns

Over the past 50 years concerns for literacy have been the preoccupation of developing countries fuelled by bodies such as UNESCO and the World Bank who have linked aid and humanitarian funding directly with human rights and economic development. It was assumed that there was a direct correlation between participation in formal education and literacy, and between literacy and economic development. It was therefore assumed that Australia did not have a literacy problem because of its long-standing policy of free, universal and compulsory education. The advocacy for adult literacy education in Australia did not really commence until the mid. 1970s, and there was very little empirical data available to support the cause. This ACAL View reports on some recent literacy surveys and what needs to happen as a result of their findings.

The issue of adult literacy problems and the need for data to inform policy is by no means new. Survey results help improve the planning, delivery and evaluation of literacy programs. Thus major national research such as quantitative surveys of needs and costs have been an instrument for policy change in Australia. We were first alerted to a potential problem by two surveys; the 1945 Army survey conducted by Arch Nelson which found 4 per cent illiteracy among the armed forces, and the 1976 Survey conducted by Judith Goyen which found significant literacy difficulties among non-English speaking immigrants in Sydney.


The Wickert Survey

The momentum of this embryonic movement was enhanced in 1989 when the first national literacy survey was undertaken by Rosie Wickert. Despite the small sample, this survey had an extraordinary political impact which fed into International Literacy Year and the subsequent development of the Australian Language and Literacy Policy

The Wickert survey was conducted at a time when similar initiatives were being proposed and conducted by a number of industrialised countries. These surveys were informed by a different definition of literacy from the literacy campaigns of the past and those associated with the Third World. Literacy was no longer seen as the achievement of a basic threshold of reading ability but as integrally linked with how adults use literacy in their everyday lives.

The impact of studies around the world, like the Wickert study, were beginning to have an effect, with many countries providing some funding for adult literacy programs. In Australia, the 1987 National Policy on Languages and the 1991 Australian Language and Literacy Policy both provided a policy framework based on these new understandings, as well as programs of action which provided the foundations of expanding provision.

There is now considerable awareness about adult literacy throughout the industrial world due, in large part, to the policy attention of the Organisation for Economic Cooperation and Development (OECD). As already indicated, this focus of attention is relatively recent. The OECD's first report about 'adult illiteracy' was published in 1992, following concerns about the results of surveys in the USA, Canada and Australia in the late 1980s (CERI 1992, p.7). Before that, illiteracy had been considered to be largely a problem for developing countries and had attracted very little attention of any kind. The OECD has now embarked on an international adult literacy survey (IALS) involving, to date, over a dozen member countries including Australia. Adult literacy, defined as 'how adults use written information to function well' (OECD 1995, p.13), has come to be seen as crucial by the OECD to both the economic performance and the 'social cohesion' of industrialised nations.


The International Literacy Survey

The IALS builds on the seminal work of Kirsch and Mosenthal (1990). Their work adopts a cognitivist approach to the identification of underlying difficulties in reading. Levine (1998, p.42-3) describes the methodology as aiming to 'establish national levels of literacy based on the capacity of samples of adults to handle everyday printed materials which require the application of literacy, numeracy and form-completion skills, in combination, to unlock the information they contain'. The findings of the international survey are classified by a range of background demographic variables. The IALS was unusual because it combined the techniques of household-based surveys with those of educational testing. In each case, the test was accompanied by a background questionnaire that allowed the interviewers to collect information about demographic and other characteristics of the respondent. Records for 45,000 individuals have been collected in the 12 countries surveyed so far.

All 101 common test items used for the assessment were open ended and taken from 'real life' situations and are intended to reflect the literacy requirements encountered in everyday life. In each of the three aspects of prose literacy, document literacy and quantitative literacy, a scale was constructed, upon which tasks of varying difficulty were placed. A person's literacy ability was then expressed by a score in each aspect, defined as the point on the scale at which he or she had an 80 per cent chance of successfully performing the task. The range of scores were grouped into five levels.

Level 1


indicates a person with very poor literacy skills.

Level 2

(226-275 points)

identifies individuals who, although they can read, can deal with only material that is simple, clearly laid out and in which the tasks involved are not too complex

Level 3

(276 - 325 points)

denotes people with the ability to integrate several sources of information and solve more complex problems. This is the level of skill regarded by many experts as a suitable minimum for coping with the demands of everyday life and work in a modern society.

Level 4 & 5

(326 -500 points)

describes respondents who demonstrate the capacity to use higher order thinking and information processing skills. Since the numbers performing at the highest skill level are small (under five percent in most countries), Level four and five are combined for the purposes of data analysis.

Australia joined the group of nations participating in the IALS in 1996 when the Australian Bureau of Statistics surveyed 9,300 people in their homes (ABS 1997). An astonishing 87 per cent of the people approached agreed to participate in the hour long 'test'. Australia therefore now has access to the best national data in its history as well as to comparative data with other OECD countries. The ABS survey provides complex information about

However before considering the findings from both the international data and the Australian results some comments about the limitations of the surveys should be made.


Limitations of the data

The Australian User Group, a committee assembled by the Australian Bureau of Statistics advised that the Australian survey could only claim to be about 'aspects' of literacy. It is significant that there were no writing tasks nor a writing scale developed. The emphasis is on information processing via reading rather than writing.

The survey only deals with English and an international view of English at that. All the tasks were developed by Statistics Canada and drawn from those countries participating in the survey prior to Australia's involvement. This means that the characterisation of some of the of everyday tasks may not consistent with the knds of ways that these things are understood in Australia. For example, many of the graphical representations were in forms not common in Australia at the time. Levine (1998) has noted that this actually means that the definition of literacy adopted was not the one that was operationalised through the items.

There is also some disquiet at the assumptions made in interpreting performance. Respondents able to complete tasks rated at a particular level for 80 per cent of the time were deemed to be at that level. There is some argument that adults will perform beyond the competence reported in the survey if they are using texts to perform tasks in a context familiar to them.

While the various components of the survey provide rich information, there is difficulty in making correlations between some of the demographic data, the performance data and the data that reports peoples' perceptions of their literacy abilities. This means that it is sometimes difficult to derive meaningful data about specific groups in the population. For example, the small size of some of the sub-samples make the data unreliable for extrapolation.

A further concern is the data at levels and 1 and 2 are more likely to be accurate than the data at higher levels. The methodology relied on increasing the complexity of tasks to make them more difficult. The performance of the higher level is far more contrived and less 'everyday' than at the lower levels, simply in order to satisfy the requirement of increased level of difficulty.

Finally, there are doubts as to the value of the data in providing a basis for long-term predictions about future requirements and challenges. More needs to be understood about literacy acquisition and changing literacy practices, and longitudinal studies are needed for this. A snapshot survey cannot tell us much about the future.

All surveys have limitations however, and they should be read as cautions in interpreting some of the results. They do not mean that nothing of value can be gleaned from such surveys. The IALS in Australia yielded much valuable information. Some of the key findings are reported in the next section.


Key findings of the Australian Survey

A part of the survey was designed to assess participants perceptions of their literacy and numeracy practices. Seventy four per cent of those surveyed do not write regularly - no more than a page once a month at the most. Twenty five per cent said they never have to write at work and sixty per cent do not read very often - perhaps once a month. Twenty eight per cent said they never have to use maths skills at work. Fourteen per cent of people stated they need help with reading information from government and business and ten per cent wanted help with filling out forms

When we look at actual performance of the survey tasks, all of which were considered to be reasonably familiar, and were in common use, the following extrapolations can be made about the levels of literacy skills of Australians aged fifteen to seventy four:

Level 1

2.6 million people

Level 2

3.6 million people

Level 3

4.8 million people

Levels 4/5

2.3 million people

Based on these and other findings, we can infer that almost half the adult population in Australia can be expected to have difficulty coping with the information processing demands of everyday life. In addition, it was reported that people who are currently unemployed have much greater difficulty that those who are in the workforce, and that literacy difficulties appear to increase with age (except for the 15 - 19 year old age group), particularly for non-English speaking migrants. Across the whole sample, of people who did not speak English as their first language, approaching half were at the lowest level on the scale. More surprising, perhaps, is that fourteen per cent of those who first spoke English are also at that level.

The Table (ABS Catalogue No. 4228.0 ) comparing the results of other countries participating in the IALS show that Australia's results are comparable. It is worth noting, however, that Sweden's results could well be a reflection of its commitments to life-long learning for all.

SEE: 'Aspects of Literacy: Assessed Literacy Skills 1996' on the Australian Bureau of Statistics website


Is It A Crisis?

Whatever critics may say about the limitations of the IALS, the findings of the adult literacy surveys are clearly a cause for concern and they are certainly an indication that this is no time to reduce funding for adult literacy and numeracy provision. The finding that almost half the population performed at levels one and two suggests that there is indeed a crisis. A crisis for the individual struggling with reading and writing in an increasingly print dominated world; a crisis for the nation as the potential of many of its citizen's remains untapped and a crisis for the economy. These kinds of findings however are not new and although they may have surprised many in the general population they would have been known to adult literacy and numeracy educators.

In spite of a number of gains since adult literacy became a political issue about ten years ago it has suffered from the hands of various administrations, partly because its clients tend to be those at greatest risk of political negotiation. There is a danger that marginal people get marginal funding. Much money has been spent and good initiatives been implemented. However programs and schemes tend to come and go, their fortunes arguably more tied to their public relations value for particular politicians than to any principled attempts to determine their successes and failures - despite the rhetoric of outcomes focus.


Policy implications

ACAL congratulates the federal government on its literacy initiatives for children, but it is disappointing that consideration of the post-school population is absent in its national literacy strategy. Australia cannot afford to ignore the literacy and numeracy training needs of up to twenty per cent of the adult population, neither can we expect them to benefit from inappropriate or knee jerk short term responses. Crisis rhetoric tends to evoke crisis responses. In adult literacy we have seen this happen too often with the result that the small gains made by participants in these programs have been lost as they get shunted from one the results of one kind of policy response to another.

The ABS survey was essential to provide both an update of adult literacy and numeracy performance as well as more fine grained analyses of particular pockets of difficulty. We cannot afford to ignore the planning possibilities that this information provides. Clearly literacy difficulties are more likely to exist within particular groups in the populations and these cannot simply be ignored. Moreover the data importantly provides further evidence that this is not a policy issue limited to school leavers or the long term unemployed. Literacy is a lifelong issue and a new national strategy is required for lifelong literacy and numeracy improvement.

The public understanding of adult literacy is growing as is the numbers of adults prepared to ask for help. The staggering response to the national adult literacy hotline provides evidence of this. So also does the success of the ABC TV literacy series, the Reading and Writing Roadshow. In light of these various indicators of need, Australia has a responsibility to develop a considered longer term policy response that sees integrated, mainstream, monitored provision - flexible, inclusive, varied, quality, and effective at achieving progress.

ACAL is supportive of nationally consistent approaches to standards development, assessment and reporting of learning outcomes. However these need to be sufficiently flexible to enable varying and appropriate responses to the diverse learning needs of different people and different contexts. Innovative approaches appear to be successful and these must not be constrained by inflexible application of quality mechanisms or reporting requirements. ACAL's experience over twenty years can help provide the necessary advice to ensure locally appropriate responses within a quality framework that allows for cultural differences.

Recent governments have consistently implemented a policy of opening up the provision of education and training to a wider set of providers with the intention of offering more choice for the user. Although ACAL has some sympathy with the notion of choice, it is not convinced that there are sufficient checks in place to ensure literacy and numeracy provision is of the quality that members of ACAL expect. There are very real challenges concerned with how to effectively build on and strengthen the considerable expertise that Australia has developed, without teachers getting caught up and compromised in a system driven by bottom line considerations. Again a longer term perspective must be developed to prevent longer term costs being shrouded by shorter term savings.

It is ACAL's view that the following need to be in place:

The Future

The costs of adult illiteracy, human, social and financial, are too great for a country like Australia to bear. Much has been achieved but these gains are at risk without a principled, strategic plan in place to secure the gains made through the initiatives of previous governments and to stop the dismantling of essential infrastructure achieved in the last ten years. There is evidence that consistent approaches pay off. The Swedish IALS results are testament to this.

A coherent articulated comprehensive national approach to adult literacy and numeracy is needed urgently in response to:

The Australian Council for Adult Literacy can and will continue to play a pivotal role in providing leadership in seeking action for adult literacy and numeracy in Australia.


Australian Bureau of Statistics (1997) Aspects of Literacy: Assessed skill levels Australia 1996, Canberra, AGPS.

Centre for Educational Research and Innovation, OECD (1992) Adult Illiteracy and Economic Performance, Paris, OECD.

Kirsch, I and P. Mosenthal (1990) Exploring Document Literacy: Variables underlying the performance of young adults, Reading Research Quarterly, 25, pp.

Levine, K (1998) Definitional and Methodological Problems in Cross-national Measurement of Adult Literacy: The case of the IALS, Written Language and Literacy, 1 (1), pp.41-61.

Organisation for Economic Cooperation and Development (1995). Literacy, Economy and Society, Paris, OECD and Minister for Industry, Canada.