Information Technology Reference
In-Depth Information
Figure 1. Fixed, mobile, and monadic usage by education, 2008
The dependent variable for the model focusing
on characteristics of broadband users by technol-
ogy is a binary (dummy) variable representing
households' broadband choices. Explanatory
variables include the age, education level, and
employment status of the main income contributor,
the number of people in the household, the number
of children in the household, the occupation of the
primary respondent, wealth, region, and habitat.
Results of the multinomial logit estimations used
to compare individuals' characteristics by the
manner in which they access the Internet for the
years 2006 and 2008 are reported in relative risk
ratios (rather than coefficients); these represent
the probability that respondents choose an alter-
native, such as mobile broadband, relative to the
omitted (reference) alternative, which in this case
is fixed broadband.
The results indicate that older persons and per-
sons with lower education levels, lower incomes,
and smaller households are the least likely to have
Internet access of any form. Madeira is the region
where people are least likely to have no Internet
access. The relative probability of a wealthier
respondent not having Internet access in the home
is 0.36, meaning most people of higher income
do have Internet access in their home.
The choice of narrowband versus fixed broad-
band follows a similar pattern, with older persons
being more likely to have narrowband than fixed
broadband. High school graduates were 2.7 times
more likely to have no Internet access than to have
fixed broadband compared to university gradu-
ates; compared to respondents with a university
degree, those who graduated from high school
were approximately half as likely to have mobile
instead of fixed broadband (relative probability
0.55). Households with at least two cars were
1.73 times more likely to get mobile broadband
instead of fixed broadband than households with
a single car or without a car.
In general, having at least one child appears to
make it more likely that a household will not have
home Internet access. Similarly, larger households
are less likely to have Internet access at home, but
household size does not affect significantly the
choice between different types of Internet access. 12
Initially the probability of not having Internet at
Search WWH ::




Custom Search