10
number of personal computers, number of schools, student enrollment,
population with junior college and above per 10,000 persons, and the literacy
rate. This study uses GDP, GDP per capita and Income of residents as the
indicators for economic factor, population with higher education and literacy rate
as the indicators for educational factor, GDP of service industry, the number of
school and student enrollment as the indicator of industry structure, and the
number of computer and telecom price index as the indicator of infrastructure for
internet adoption, population and urban population as the indicator of population
structure. Statistics on populations’ English skills is not available; this study
assumes that the high educational level indicates the higher command of English
skills. In China, children received compulsive education will generally learn
English from 5
th
or 6
th
grade in elementary school and the learning continues into
college. Since telecom market still remained monopoly control for the whole
nation in year of 2000 and 2001, the factor of telecom competitiveness is not
considered in the model.
The percentage of Internet users by geographical distribution was
provided by CNNIC survey (Jan. 2001). Multiplying this percentage with the
national population (2000), we can obtain the number of Internet users in each
province/municipality.
Findings
1. Diffusion curve
To examine if a diffusion process fits a normal S curve, first a stable
social-economic environment free from change and in which everyone has same
Convention