Page 126 - Culture Technology Communication

National Level Culture and Global Diffusion 109

Table 5
Reduced Sample Models for START
Model N Adjusted R 2 R 2
Model 1 55 .454
TELEDEN .444 .580 .01
TOEFL .030 .194 .1
Model 2 55 .546
INTCALL .444 .512 .01
SCHLNROL .119 .378 .01
Model 3 55 .435
GDP_CAP .345 .280 .01
SCHLNROL .086 .358 .05
TOEFL .035 .210 .1
Model 4 55 .516
INTCALL .444 .517 .01
NWSPAPR .090 .335 .01
Model 5 55 .364
SCHLNROL .345 .469 .01
FAX1993 .043 .238 .1
Model 6 55 .356
GDP_CAP .329 .464 .01
TOEFL .051 .251 .05
Model 7 55 .394
SCHLNROL .345 .388 .01
NWSPAPR .072 .334 .05

model 7 shows that in a direct comparison with School Enrollment,
it has less explanatory power.

GROWTH IN INDIVIDUAL COUNTRIES
The final part of this analysis is concerned with the factors influenc-
ing growth in individual nations. International diffusion studies can
be concerned with breadth, the diffusion across countries, or in its
depth, the diffusion of the innovation within each country. When ex-
amining depth the dependent variable will be a measure of growth.
Not all countries have had the Internet for a sufficient length of
time to judge their growth. By first taking those countries with ten
or more data points and removing outliers, a sample of forty-eight
countries was created. These countries, because they are earlier
adopters of the Internet comparatively, do not constitute a represen-
tative sample of the countries used in the analysis above. In fact,
they are expected to score higher on the variables found to be signif-
icant predictors of adoption timing (lower international call costs,