20. New Zealand has a high level of connectivity to the Internet, ranking fifth in the top ten OECD countries by hosts per 1000 people.⁷ By February 2000 there were 48,230 registered domain names in the .nz domain. If we assume one domain name per organisation, this gives the approximate number of organisations connected to the Internet. Commercial domain name registration is growing at an increasing rate.

21. While overshadowed by the numbers of commercial domains, growth in the registration of non-commercial domains has increased by 76 percent in the twelve months to February 2000.

22. Organisations with Internet registered domain names do not necessarily have Web sites on the Internet.⁸ It is estimated that as at February 2000 there were 25,342 Web sites in New Zealand of which 87 percent have been set up by commercial organisations. Overall there has been an increase of 75 percent in Web sites over the year from February 1999. Some commercial web sites are registering as dot com, such as 7am.com, the international news service run from the Kaipara Harbour.

23. The 1998 Lincoln survey of farmers found that 28% of respondents with computers had an Internet connection. The recent MAF research found that 81% of respondents with computers had an Internet connection. This illustrates the rate at which farmers are getting online, although as noted previously, there are problems with the technical infrastructure that could hamper this. For example, 69% of respondents indicated that the speed of their Internet access needed to be faster.

24. A number of other studies that look at Internet access, and how it differs for different demographic groups, have been conducted in the past couple of years. These surveys are summarised in the next few sections.

Internet access - AGB McNair survey, 1998

25. An AGB McNair survey (probably 1998) found that 81 percent of Internet users were European, 6 percent Maori (15% of population), 3 percent Pacific Island peoples (6% of population) and 10 percent other ethnicities (5% of population). The same survey found the largest occupational group of users is students, followed by trade/technical professionals and managers.

Internet access - AC Nielsen, December 1998

26. A December 1998 study on Internet and Computer usage in New Zealand found that around 30 percent of people had accessed the Internet in the month preceding the study⁹. The study showed that overall variation in the likelihood of having accessed the Internet in the past four weeks was best explained by age, household income and work status.¹⁰ Users were generally younger, in higher income brackets and in work. Just over 15% of people earning under $30,000 had ever accessed the Internet (despite this group having a high concentration of students, who do tend to be heavy users), compared to over 30% of those earning between $30,000 and $60,000. This compares with nearly 40 percent of those earning between $40,000-$80,000, and nearly 60 percent for those earning over $80,000.

27. Males were more likely than females to have ever accessed the Internet. While home was the most common place for accessing the Internet, females were more likely than males to gain access through school/tech/ university while males are more likely than females to gain access at work.¹¹

Internet access - AC Nielsen Xtra survey, late 1999

28. The aforementioned AC Nielsen Xtra survey reported the following information:
    • Internet usage in last 4 weeks 34% (up 42% on a year previously)

    • Internet usage by gender: Male 38%, Female 31%

    • By Age group: 10-19 48%, 20-29 44%, 30-39 40%, 40-49 36%, 50-59 28%, 60+ 9%

    • By household Income: >$20K 13%, $20-30K 19%, $30-40K 24%, $40-60K 37%, $60-80K 47%, $80-100K 58% , $100-120K 64%, $120K+ 78%

    • By city: Auckland 39%, Wellington 46%, Christchurch 44%, Dunedin 31%, Hamilton 30%

    • Service used: e-mail 82%, www 80%, news groups 12%, chat 12%

    • Top activities last 4 weeks: General surfing 56%, Info products/services 38%, Info co/Org 35%, News/current affairs 32%, Download software 26%.

29. The sample sizes of the AC Nielsen surveys were insufficient to produce robust estimates of access by ethnicity.

Internet access - NRB survey for Porirua City Council, December 1999

30. Ethnicity and income differences were also found in a recent survey on another topic carried out for Porirua City Council that included a question on access to the Internet.¹² While 50 percent of European/other residents had access, only 17 percent of Maori/Pacific Island residents had access.¹³ While 76 percent of those earning more than $50,000 had access, this dropped to 27 percent for those earning between $25,000 - $50,000, and to 15 percent for those earning less than $25,000. This reinforces the overseas data that income and ethnicity are primary factors in relation to access to the Internet. This also results in a geographical divide in relation to access: while 63 percent of residents in one ward had access, this contrasted with 37 percent and 17 percent access respectively in the two other wards.

Internet access - AC Nielsen/NetRatings, 2000

31. Relative to other developed countries, New Zealand seems to have a relatively high rate of Internet usage. An AC Nielsen comparison of the United States with Australia, New Zealand, Ireland, the United Kingdom and Singapore¹⁴ found that of an estimated 1.7 million New Zealanders with Internet access, around 565,000 had gone online during March 2000 (47%), less than the US and Australia, but more than the UK and Singapore. In terms of average time spent online per month, New Zealand ranked second only to the US. Males represented 55% of the New Zealand Internet audience, and the 35-49 age group was the most active (around 30%).

Internet access - Ministry of Social Policy standard of living survey, 2000

32. The Ministry of Social Policy (MSP) has recently funded a national survey focusing on the standard of living of working age people. The data is still at an early stage of analysis, but responses to questions concerning Internet access are useful in relation to this stocktake and have been subjected to a first-cut analysis, using age, ethnicity and gender. The surveys also collected other socio-economic and demographic data: it will be possible at a later date to relate patterns of preference, ownership and use, to key variables such as income and education that international research suggests influence the digital divide.

33. A series of questions asked respondents to identify their ownership and preferences for various consumer goods and services. These included personal computers and Internet access. Information was also collected on whether barriers were related to costs or other reasons.

34. The following table shows the distribution of Internet access by several other variables (age, PC ownership, ethnicity, gender). Respondents not having access were asked whether they wanted it or not and, if so, what prevented them from having it (cost and non-cost barriers).

Working age people - percentages within age groups by category of Internet access
Category	Have Internet access	Want Internet access and don't have it because of cost	Want Internet access and don't have it for other reasons
18-34 35-49 50-64	39.0% 53.6% 40.9%	25.8% 15.1% 10.7%	13.1% 9.6% 12.0%
Own a PC	73.6%	8.9%	9.1%
European Maori Pacific peoples Chinese & Indian	48.2% 26.0% 12.4% 64.4%	15.6% 32.8% 36.5% 17.6%	12.4% 10.6% 9.1% 5.1%
Male Female	46.5% 42.7%	16.6% 20.8%	12.4% 10.8%
All groups	44.6%	18.7%	11.6%

35. Looking first at age, there are three striking - and more-or-less predictable - patterns in the responses. The first is that, in the year 2000, the overall preference for Internet access appears to be high in the New Zealand population aged between 18 and 49 years, but declines rapidly amongst 50-64 year-olds. It is expected that this pattern continues into the 65+ age group, a point to be examined using other MSP survey data that will become available next year. Access is presently at its highest amongst those aged 35-49 years - probably linked to discretionary income and to family-related motivations.

36. A second pattern in the responses relates to reasons for non-access. For all ages up to 50 years, cost is the single most important barrier and clearly represents a significant element of the digital divide. This is most evident for 18-34 year olds where cost excludes a quarter of the population. Non-cost reasons were not explored in the survey. Analysis in the future will examine the relationship between income, education and lack of access apparent from these data, but it is probable that a significant proportion of those excluded are those on lower incomes. Given that these people are probably the highest users of government social services, this implies significant implications for the form and rate at which e-government is introduced over coming years.

37. The issue of cost can be seen in more detail by looking at the responses of all 18-64 year olds to questions regarding ownership of computers and Internet access. Generally, as would be expected, ownership of a computer is strongly associated with Internet access - although, because of household, family and community relationships and workplace facilities, some individuals have Internet access without owning a personal computer.

38. Notably, almost 10% of working age people who own a personal computer are unable to afford the connection and ongoing line charges associated with Internet connection. Around the same proportion have non-cost reasons for not being connected. In all, around 30% of the 18-64 year old population have a PC and yet are still subject to the digital divide.

39. The ethnic dimensions of Internet access are also revealing. Overall, 76 percent of European/Pakeha respondents of working age indicated that they have or want to have Internet access. The equivalent figures for New Zealand Maori were 40 percent, 58 percent for Pacific peoples and 87% for Chinese and Indian respondents, with the large majority of this last group already having access. But the proportions of those unable to purchase Internet access because of cost factors - 32.8 percent for Maori and 36.5 percent for Pacific peoples - is significantly greater than those who have purchased it. This probably reflects the tendency towards lower income levels for these groups. The important point though is that there is a level of aspiration amongst these groups to which price is a definite barrier.

40. There are often assumed to be gender differences in Internet access but the patterns evidenced in the above table challenge that assumption. The MSP data reveals no major differences between males and females of working age in terms of Internet access. Future, more detailed analysis of the full data set may throw up significant disparities for sub-groups of females, or differences between coupled/single males and females, but this overall population data indicates that in general, females are as connected as males.

Access issues for communities and community groups

41. Some countries have established Community Technology Centres as a way of providing poorer individuals and communities with access to ICT. In New Zealand however currently there is no integrated strategy for ensuring that there are adequate public ICT access points in low income areas, either commercially or community operated. Officials report that what community access exists has developed organically and is likely to suffer from inadequate funding and technical support.

42. While no comprehensive data are available it appears that many community organisations see the benefits both to them organisationally and to the communities they serve, of being "wired"¹⁵. However it seems that many are struggling to improve their ICT connectivity, for reasons of cost and lack of skill/technical support.

43. Feedback from Child, Youth and Family suggests a wide range of access for community organisations. Some organisations had no electronic equipment, other than telephones and faxes, while others have quite sophisticated, connected data systems. Many of the larger organisations have quite sophisticated technology and would appreciate undertaking more of reporting on-line. It would ease their workload and make operations more efficient.

44. As part of their community online strategy the Department of Internal Affairs (DIA) sponsored the Flaxroots Technology conference in April this year, which generated a lot of interest from community groups.¹⁶ Feedback made to a survey of the CommunityNet Aotearoa website, and discussion on the Community Networking New Zealand email group has included the following points about access:
    • "There is a need to promote a general government policy about the right of small communities to have free internet access. We are in a provincial area and whilst we actively promote community net, there are many groups without access to PCs.

    • "There is a need for specialised assistance such as ….. a strategy to provide computers in all community centres, grants for acquiring second hand computers, exemptions from software licensing for voluntary organisations, …. an online facility to enable cooperative purchasing of fixtures, fittings and equipment, a nation wide register of successful projects…"

    • from Te Araroa "...we hope to provide Internet access in our [information] centre, the barrier is money shortage".

    • "….[Policies], websites and other Internet resources are made for people with PC's less than three years old, running Windows98 or later, Internet explorer 4 or 5, with 56K modems or above on reliable connections, cheap or free phone calls and cheap and accessible IT support skills."

45. Through their interaction with people in community organisations Community Development Group (CDG) Advisors are aware of the reasons why communities are not getting online. As one advisor commented:

    "More and more I come across groups that: have their own internet access, have at least 1 committee member, with internet access or have access to their web mail through public libraries, friends, family members, uni polytech or high school computer suites. However, generally speaking there are still lots of groups without their own pc and internet access, or access full stop.

46. Access barriers advisors have come across in the community sector are inability to purchase equipment, high cost of ISP and telephony requirements, technical problems such as line quality and installation, and the high cost of upgrades.

47. Community groups' use of the Internet still appears limited. In the central North Island, of 370 community sector applicant groups to Community Organisations Grants Schemes (COGS) committees covering Hauraki, Western Waikato, Hamilton City and South Waikato, only 84 organisations have e-mail. Out of the 141 grant applications received by the Waitakere COGS Committee, only 45 of the groups had an e-mail address and approximately half of these were the e-mail addresses of individuals in the groups, rather than a "dot org" address. In Kaitaia, only 32 out of 115 COGS applicant groups had an email address.

48. Applicants surveyed as part of the Lottery Grants Board annual customer satisfaction survey 1999-2000¹⁷ have provided a picture of community Internet access. (Sample size = 456, with a margin of error ± 4.6%.) Numbers of applicants with an email address: 53%. The level of access varied according to which Committee an application was made to:
    • Lottery Health Research (98%)

    • General (68%)

    • Welfare (55%)

    • Youth (54%)

    • Lottery Community Facilities (50%)

    • Environment & Heritage (40%)

    • Aged (39%)

    • Marae Heritage (35%)

49. Significantly higher levels of access are recorded amongst applicants making applications in excess of $40,000 (68%) and organisations with turnover of $100,000 or more (77%). Significantly lower than average levels of email access are recorded amongst those applying for under $2,000 (31%), those who became aware of Lotteries through the newspaper (33%), and organisations with a turnover of up to $15,000 (23%)

50. Fifty two percent of applicant organisations had an Internet connection. The level of access follows a similar pattern to that of applicants having email addresses. It is noted that "[t]here is a high correlation between applicants that have an internet connection and those who have email (91% of applicants that have email also have an internet connection). Thus, the demographic profiles of both groups are similar."

51. To illustrate the limited ability of not-for-profit organisations to exploit the value of the Internet, the National Collective of Women's Refuges has only six out of fifty refuges connected by e-mail.