| Prose literacy level | |||
|---|---|---|---|
| Levels 1 and 2 | Level 3 | Levels 4 and 5 | |
| Bermuda | 1,00 | 2,38** | 3,39** |
| Canada | 1,00 | 1,83** | 2,40** |
| Italy | 1,00 | 1,64* | 2,02* |
| Norway | 1,00 | 1,80** | 2,66** |
| Switzerland | 1,00 | 1,78** | 1,91* |
| United States | 1,00 | 1,59** | 1,94** |
Note: In order to produce reliable estimates, Levels 1 and 2 and 4 and 5 have been collapsed. Levels 1 and 2 are used as the reference group. Odds are adjusted for gender, age, educational attainment, labour force status and total household income.
* statistically significant at the 5% level.
* * statistically significant at the 1% level.
Source: Veenhof, Clermont and Sciadas, 2005.
While ICT penetration and use have been analyzed extensively, more research is needed to understand
outcomes associated with such use. Such an effort becomes more illustrative if ICT use is combined with literacy skills. This section represents an attempt to assess such outcomes, by exploring the association between personal income and a combined measure of literacy skills and ICT use.
As in the previous section, a logistic regression model is used to estimate the odds of being a high-income
earner (dependent variable), while controlling for various socio-economic characteristics. In this case, groups
with different profiles of literacy skills and computer use are included in the model (see Box 4 for details regarding the delineation of the groups).
BOX 4: Combined literacy and computer use profiles
The logistic regression in this section models the effects of various socio-economic characteristics, as well as
literacy and computer use profiles on personal income. Respondents were divided into 4 groups on the basis of
their literacy and computer use profiles as follows:
Group Prose literacy level Level of use of computers for task-oriented purposes
Group 1 below average (levels 1 and 2) low-to-medium intensity (lowest 75% of computer users)
Group 2 average or higher (levels 3 to 5) low-to-medium intensity (lowest 75% of computer users)
Group 3 below average (levels 1 and 2) high-intensity (top 25% of computer users)
Group 4 average or higher (levels 3 to 5) high-intensity (top 25% of computer users)
The regression estimates the odds of being in the top income quartile (highest 25%) of personal income,
relative to the reference group (Group 1).