This article replicates previous findings that the diffusion of the Internet is becoming more polarized by family income in the United States. Using multiple logistic regression and other odds-based analyses to assess Internet access in the United States from 1998 to 2001, the recent analysis confirms that the odds of access increased most rapidly for individuals at highest family income levels, and most slowly for individuals with the lowest income levels. The unique divide in the U.S. is further evidenced by the lack of such static and dynamic income differences in the U.S. compared with income differences in data from 15 European nations.

Moreover, such application of odds-based measures of Internet diffusion is supported by the relative lack of differences in comparisons by education, age and other demographic variables (besides income) for the U.S. Issues about the assumptions underlying use of odds-based measures are discussed.

This article proposes a relatively simple indicator, DIDIX, that was developed in an EU-funded project to benchmark and track national digital divides within EU member states. Based on the relative diffusion of computers and the Internet in four disadvantaged socio-demographic groups (compared to national averages), the index is intended as a descriptive metric to compare basic levels of inclusion in EU-member states.

Because it reaches lower levels at higher rates of diffusion, however, DIDIX is neither proposed as a way to identify diffusion patterns at an early stage, nor to predict future developments. Its intent instead is to compare the diffusion of technology in at-risk groups with the population average. Results here suggest an increasing North-South gradient of cross-national inclusion prevailing in Europe. Applying the underlying methodology to other than simple access or use variables suggests that more attention should be paid to indexing the various skills and general benefits of IT.

The diffusion of the Internet (and its accompanying digital divides) has occurred at the intersection of both international and within-country differences in socioeconomic, technological and linguistic factors. Telecommunications policies, infrastructures and education are prerequisites for marginalized communities to participate in the information age. High costs, English language dominance, the lack of relevant content, and the lack of technological support are barriers for disadvantaged communities using computers and the Internet.

The diffusion of Internet use in developed countries may be slowing and even stalling, when compared to the explosive growth of Internet access and use in the past decade. With the proliferation of the Internet in developed countries, the digital divide between North American and developed countries elsewhere is thus narrowing, but remains substantial. The divide also remains substantial within almost all countries, and is widening even as the number and percentage of Internet users increases, as newcomers to the Internet are demographically similar to those already online. People, social groups and nations on the wrong side of the digital divide may be increasingly excluded from knowledge-based societies and economies.

By virtue of their physical make-up, their cultural and linguistic diversity, and the relative isolation and spread of their population, Pacific Island countries are faced with a multitude of challenges in the delivery of information services. This article reviews the nature of the digital divides that exist in the Pacific region, considering divides within countries, between the countries, and between the Pacific region and the rest of the world.

The varied but generally high costs of Internet access (in part brought about by national telecommunication monopolies) are exacerbating the digital divide along socio-economic lines; but they also create regional imbalances, with certain countries effectively isolated. Nonetheless, community-based systems can work to offset this, as shown on Niue. Within these countries at present, no structures are even envisaged that would address digital divides, nor the implications of the technologies on traditional rank, status and power structures, which are fundamental matters in Polynesian and Micronesian societies.