TABLE A1
Logistic Regression Table
Internet Access with Key Demographics
| |
In the past three months |
At home |
| exp (B) |
sig. |
exp (B) |
sig. |
| 1999 |
| Education |
1.3141 |
(+) |
1.1822 |
(+) |
| Age |
0.6542 |
(-) |
0.806 |
(-) |
| Income |
1.4115 |
(+) |
1.4191 |
(+) |
| Sex |
0.7412 |
(-) |
0.7926 |
(-) |
| Rural |
1.3076 |
(+) |
1.1059 |
n.s. |
| 2000 |
| Education |
1.3218 |
(+) |
1.2163 |
(+) |
| Age |
0.5709 |
(-) |
0.7085 |
(-) |
| Income |
1.3794 |
(+) |
1.4019 |
(+) |
| Sex |
0.7722 |
(-) |
0.6854 |
(-) |
| Rural |
1.4195 |
(+) |
1.2209 |
(+) |
note:
Sex and rural are categorical variables
Sex: 0 = men; 1 = women
Rural: 0 = rural; 1 = urban |
In the 2000 survey, the results show that
the five key demographic characteristics remain in a very similar pattern.
The urban/rural setting gains in importance over income and education
(all positively correlated) while the age variable becomes increasingly
important in determining the likelihood of recent on-line access (negative
correlation).
The likelihood of having access to the Internet at home (a sub-sample
of recent Internet users) reflects obvious similar patterns to the abovementioned
results. In 1999, however, the location (urban or rural) does not emerge
as a significant predictor. It re-appears, nonetheless, in the results
from the 2000 survey.
|