• Summary
• Downloads
• Explanatory Notes
• Related Information
• Past Releases

18/01/2008 Note: Data Cubes 1 - 25 are being reissued to include the correct units ('000 and %) in the column headings. There is no change to the data.

NOTES


INQUIRIES

For further information about these and related statistics, contact Siddhartha De on Canberra (02) 6252 6519 or the National Information and Referral Service on 1300 135 070.

SUMMARY COMMENTARY


CHAPTER 1: MAIN FEATURES

There has been significant growth in Australia's access to/use of the Internet between 2001 and 2006. In 2001, 35% of Australian dwellings had access to the Internet in the week prior to the Census date. In 2006, 63% of Dwellings had access to the Internet.

1.1 Regional differences in access

Both cross-tabular and regression analyses reveal considerably lower access rates for regional and rural areas, in comparison with major cities of Australia, especially for Broadband access.

At the national level 66% of dwellings in major cities have access to the Internet, compared to 42% for very remote Australia. This gap is similar for Broadband access, the corresponding figures being 46% and 24%. Corresponding access rates for Inner Regional, Outer Regional and Remote Australia are 56%, 52% and 53% for Internet access and 32%, 27% and 28% for Broadband access.

Regression analysis results reveal that regional and remote areas are at least 40% less likely to have Broadband access relative to major cities. The likelihood of any Internet access is relatively higher, but still considerably lower than major cities.

In respect of States and territories, considerable differences in access rates were recorded, both for any Internet and Broadband. The Australian Capital Territory has the highest proportion of occupied dwellings connected to the Internet (75%). New South Wales, Victoria, Queensland and Western Australia have similar levels of Internet connection, ranging between 63% and 65%. Likewise, South Australia, Tasmania and the Northern Territory have similar levels of connectivity, ranging between 55% and 58%. Similar patterns were observed for Broadband connectivity as well. The Australian Capital Territory has the highest proportion of occupied dwellings having Broadband connectivity (53%). New South Wales, Victoria, Queensland and Western Australia have very similar levels of Broadband connectivity, ranging between 41% and 42%. South Australia, Tasmania and the Northern Territory have similar levels of connectivity, ranging between 28% and 32%.

1.2 Socio-economic

Income

Rates of access continue to vary significantly with income. Based on equivalised household incomes, individuals living in households with equivalised income of $2000 or more per week are three times more likely to have Broadband access compared with persons with less than $600 per week income. The results are in line with recent related ABS surveys such as 2005-06 Household Use of Information Technology (HUIT), in which 22% of households in the bottom two equivalised income quintiles stated high cost as the main reason for not having Internet access, and only 34% of people in the bottom income quintile households had home Internet access, compared to 77% in the top income quintile.

Educational attainment

Educational attainment is another factor influencing any Internet access. For example, regression results indicate that, compared to people with no post school qualifications, people with post graduate degrees were 83% more likely to have Broadband access.

Family composition

Families (both couple and single parent) with children under 15 and dependent students are most likely to be connected. Regression results indicate that such families have a three to four times higher likelihood of Broadband access in comparison with families without children or dependent students, signifying the importance families with students assign to Broadband connectivity.

Labour force status and occupation

Unemployed people are 12% less likely, and people not in the labour force are 18% less likely, to have access to Broadband in comparison with employed people in high skill occupations. People employed in low skill occupations are 27% less likely to have Broadband access.

Gender and marital status

Both unmarried males and females are less likely (by 25% and 37% respectively) to have Broadband access than married males and married females.

English proficiency

In contrast to previous studies, people with no proficiency in English are slightly more likely (6%) to have Broadband access than people proficient in English. People with poor proficiency in English are 27% less likely. The results are slightly different for any Internet access, with people with no English proficiency being 8% less likely to have access.

Age

In comparison with people aged between 35 and 44 years, people under 24 years of age have more than 50% likelihood of Broadband access. Older people are less likely to have Broadband access, people in the age group of 65 to 74 being 42% less likely, and those 75 years or more 34% less likely.

Indigenous status

Indigenous people are about half as likely to have Broadband access compared to non-indigenous people.

Disability

Based on the results of cross-tabular analysis, only 28% of people requiring assistance with core activities have Broadband access, in comparison with 48% for people not needing assistance.

A more complete analysis of each of these characteristics is provided in the body of this publication, together with further fine level detail in the attached Appendix maps and tables.

CHAPTER 2: INTRODUCTION

Use of the Internet as a conduit for communicating, accessing information and undertaking commerce has increased significantly in the last decade. World wide, people using the Internet is expanding every year, and Australians are embracing this technology rapidly. Research in Australia based on available data suggests that significant differences exist in Internet access based on income, education and age. Although it is predicted by some experts that the Internet will soon become as ubiquitous as television, there are concerns that in addition to costs relating to purchasing equipment and Internet access, lack of cognitive ability and technical skills to exploit the Internet could lead to a social divide (Curtin, 2001).

There has been a strong policy focus as well as community interest in recent years on access to, and use of, the Internet, especially Broadband. Policy makers are increasingly focussing on the economic and social impacts arising from the adoption and use of Information and Communications Technology (ICT). This study, following on similar studies based on the 2001 Census data, identifies key socio-economic characteristics of households and individuals relating to access to the Internet using results from the Census of Population and Housing (hereafter referred to as "the Census") 2006.

The use of the Internet by individuals depends on many factors including where the Internet is accessed (at work, school, home or in other locations), how affordable access is, and the ability and interest of users. Businesses and government agencies are keen to understand the characteristics of users. This supports targeting of potential customers and facilitating service delivery. Governments, community organisations and social researchers are interested in understanding the barriers to using the Internet, with a view to assessing the degree of exclusion from the information society and its impacts on social and economic outcomes. Policy makers target policies to address inequities in access.

The Census 2006 provides a wealth of information for investigating the socio-economic and regional characteristics relating to Internet access by households. Several research and information papers such as the Australia on-line: How Australians are using Computers and Internet 2001 (ABS, 2001) and National Centre for Social and Economic modelling (NATSEM) report series based on Census 2001 data have been published on this topic. In this study, the 2006 Census data relating to access to the Internet in Australian cities and regional households is used for a detailed analysis of this issue from a regional as well as socio-economic perspective.

2.1 Structure of Paper

The paper has been structured into 5 broad sections:

• Section 1: Main Features and Introduction
• Section 2: Geographic aspects
• Section 3: Socio-economic aspects
• Section 4: Combined analysis
• Section 5: Conclusions

Appendices to this study, available on the ABS web site (www.abs.gov.au) include thematic maps depicting Internet connectivity at various levels of detail to the CD level , as well as tables of data relating to the study.

2.2 An Overview of the Previous Work

The Census of Population and Housing 2001 included two questions about the use of computers and the Internet by individuals in the week prior to the Census night. The Internet question also included the location of use. These questions provided the data for several research studies relating to Internet use in Australia. A selection of these studies are summarised in the following paragraphs.

Using the 2001 Census data, Lloyd and Bill (2004) examined the socio-economic and regional characteristics of users of home computers and the Internet in Australia, and found significant variation in rates of Internet use based on socio-economic characteristics of users. People with higher incomes and better education were found to be more likely to access the Internet. Families with dependent children were more likely to be Internet users compared to those families without children. Labour force status was another major factor influencing the use of Internet by individuals - employed people were more likely to be users compared with the unemployed or people looking for full-time work. For example, in 2001 people classified as unemployed and looking for part-time work reported the highest rate of Internet use at home.

Age, English language proficiency, indigenous status and country of birth also had a significant influence on Internet use. Older persons, especially older women, recorded lower rates. People who do not speak English at all, who did not go to school beyond year 8, Indigenous people and those born in Southern and Eastern Europe were found to be the groups recording significantly lower rates of Internet use in 2001.

In another study using the Census of Population and Housing in 2001, Daly (2005) highlighted the low levels of Internet usage by Indigenous Australians, and concluded that lower rates of use was influenced mainly by lower levels of income and the education of Indigenous people.

In addition to these studies based on the Census of Population and Housing 2001, there have been studies based on survey data using the ABS HUIT Survey (1998 and 1999) and KPMG Household Survey (2000). Hellwig and Lloyd and Hellwig (2000) and Lloyd et. al. (2000) reported that educational qualification was the major driver of the Internet use by individuals, followed by income. They also found that people receiving government benefits were more likely to have Internet access. The study concluded that disadvantages were more likely due to socio-economic influences rather than geographic barriers. Xavier (2001) noted evidence from other countries that socio-economic factors such as income, level of education attainment, gender, age, and disabilities are major determinants of Internet access and usage patterns.

Based on a literature review Curtin (2001) concluded that digital technology per se has not created a new social divide. Before the Internet could be heralded as an egalitarian medium, a range of social, economic and technological barriers needed to be addressed. According to Curtin, in 2001 many Australians had 'reasonable' access to the Internet. A small percentage living in rural and remote Australia in particular, had very limited access.

McLaren and Zappala (2002) concluded that despite figures suggesting that Australia is a high consumer of ICT goods and services, the consumption was not spread evenly across the population. This study was based on around 3000 households and 6000 children from financially disadvantaged backgrounds. This study also found that there was a strong association between the level of parental education and ICT access and use.

In a study on barriers to e-learning, the Australian Institute for Social Research identified that people with low incomes, people who do not have tertiary level education, people who live in rural and remote areas, people of Aboriginal or Torres Straight Islander heritage, people with disability, people of non-English speaking background, unemployed people, people who are aged over 55 years, and women are groups that are under-represented in terms of Internet connectivity and use of ICT (The University of Adelaide, 2006).

2.3 Data sources

The 2006 Census of Population and Housing is the primary data source for this study. The Census is a valuable data source for identifying the size and geographic distribution of the Australian population, and for analysing the major demographic, social and economic characteristics of the population, particularly for small geographic regions and other small sub-populations. It provides statistics for decision-making by governments, businesses, community organisations and individuals (ABS, 2007).

The Census was designed to measure the number of people in Australia on the Census night (8 August 2006), their key characteristics, and the characteristics of the dwellings in which they live. The count excluded foreign diplomats and their families. Overseas visitors, although enumerated in the Census, have been excluded from this analysis. Australian residents outside the country on the Census night are also excluded.

The analysis of Internet access is based on occupied private dwellings only. It excludes dwellings which have not responded to the Internet access question. There are 7,028,870 such dwellings in scope of this analysis, out of a total 7,596,182 private occupied dwellings. If it is assumed that non-responding dwellings do not have Internet access, there could be a corresponding upward bias in the estimated connectivity levels of about 7% at the national level. However, this study is based on reporting households only and does not make adjustments for potential non-response bias.

As a result, some 2006 Census classifications are different from their 2001 counterparts. Where the changes are significant, such as in the case of adopting new classifications in the Census, the ABS has developed concordances to assist users to compare Census data over time (ABS, 2006b). There are a number of conceptual and classification changes to be applied to the 2006 Census. These changes are provided in full detail in the 2006 Census Dictionary (cat. no. 2901.0).

The 2006 Census Dictionary is a comprehensive reference guide designed to assist users of 2006 Census data to understand the concepts underlying the data. The Glossary section of the dictionary defines the Census terms used in this publication.

2.4 Comparison with results from 2001 Census

Both the 2001 and 2006 Censuses included questions relating to the Internet. The 2001 Census included two simple questions relating to use of computers and the Internet (along with location of Internet use) by individuals in the week prior to the Census night. The Census 2006 question related to Internet access, as well as type of access, in dwellings. Therefore, information available from the two Censuses is not directly comparable. However, a close match for accessibility was achieved for 2001 Census Internet question by assuming use at home by individuals equating to dwelling Internet access. With technologies such as mobile Broadband not being in existence in 2001, this assumption is considered to be realistic. It should be noted that this assumption will cause some undercount of the number of dwellings with Internet access, for situations where the dwelling had Internet access but use was not made in the week prior to the Census night.

The Census 2006 question was based on Internet access in the dwelling. For the purpose of this study, it had been assumed that if the dwelling had access to the Internet, any person living in that dwelling had access to the Internet. It should be noted that this assumption leads to an overestimation of the number of persons actually accessing the Internet in that dwelling, especially for population groups such as the elderly, the disabled and less educated.

2.5 Comparison with results from related ABS Household Surveys

The analysis also covers comparison with the broad results of the HUIT Survey for 2005-06. HUIT data is collected from the Multi-Purpose Household Survey (MPHS). The MPHS is conducted each year throughout Australia from July to June as a supplement to the Monthly Labour Force Survey, which is designed to collect statistics for a number of small, self-contained topics, such as Internet use. It should be noted that HUIT data is collected over a 12 month reference period while the Census data relates to a specific night. Therefore comparisons should be made with caution.

2.6 Summary of changes to major classifications

Some classifications used for grouping the Census data have undergone major revisions since the last Census. Changes to major classifications are listed below.

The first edition of the Australian and New Zealand Standard Industrial Classification (ANZSIC) (cat. no.1292.0) released in 1993 was used to classify responses to Census questions on Industry of Employment for the 1996 and 2001 Censuses. The second edition (2006 revision) of the classification is used to output standard tables for the 2006 Census (2006b).

The Australian Standard Classification of Languages (ASCL) (cat.no.1267.0), is used to classify the variable Main Language Other Than English Spoken at Home. The classification has been revised since the 2001 Census (ABS, 2006b).

The Australian Standard Classification of Occupations, Second Edition (cat. no. 1220.0), was used to classify responses to Census questions on Occupation for the 2001 Census. The new Australian and New Zealand Standard Classification of Occupations (ANZSCO) (cat. no. 1220.0), is used to output standard tables for the 2006 Census (ABS, 2006b).

Ancestry is coded using the Australian Standard Classification of Cultural and Ethnic Groups (ASCCEG) (cat.no.1249.0). The ASCCEG has had a minor revision since the 2001 Census (ABS, 2006b).

The geographic classifications are based on the Australian Standard Geographical Classification (ASGC) (cat.no.1216.0) 2007. The ASGC provides a common framework of statistical geography. The Main Structure, the Statistical Region Structure, the Section of State Structure, and the Remoteness Structure cover the whole of Australia without gaps or overlaps.