controlling for other contextual and individual-level factors. Individual-level demographic
factors as well as concentrated poverty define the contours of the digital divide in terms of access
to a home computer. While previous research has focused on the individual-level predictors of
access to information technology, no previous research we are aware of has shown the
importance of concentrated poverty and the environment on access to technology.
The educational attainment of the community is also important as shown in baseline
model 2 (Table 2). Respondents residing in zip codes with lower educational attainment
(measured by percent of the population with a high school diploma or higher) are statistically
less likely to have a home computer than those residing in geographic areas with higher
educational attainment. This suggests that socioeconomic context (and possibly educational
opportunities) measured by educational attainment of a community is important in understanding
digital inequalities. In sum, individuals residing in low socioeconomic status environments
(measured by median income or educational attainment) have statistically lower probably of
access to a home computer.
In both baseline models, Latinos had lower access rates than other racial and ethnic
groups, but in baseline model 2, residents of zip codes with higher Latino populations had
statistically higher access rates. How can this be? These data indicate that other cultural or
regional characteristics of geographic areas with large Latinos populations are associated with
increased computer use. Latino populations are concentrated in the southwest, a region with the
highest percent of high-technology industries. The percentage of African-Americans or Asian-
Americans in the respondent’s zip code had no measurable impact on home computer access.
The question driving this research, however, is the interaction of race and place on
technology access. What is the effect of being a racial or ethnic minority and living in a poor
community on the probability of access to a home computer? Hierarchical linear models
presented in columns 4 and 5 (Table 2) are identical to those presented in columns 2 and 3, but
19
Convention