Computer literacy, employment and earnings : a cross-sectional study on South Africa using the National Income Dynamics Study 2008.

In this study I explore the extent of computer literacy in South Africa, the correlates of computer literacy, and the relationship between computer literacy and labour market outcomes, namely the probability of employment and earnings among working-age South Africans. I use data from the first wave...

Full description

Bibliographic Details
Main Author: Govindasamy, Preston-Lee.
Other Authors: Casale, Daniela Maria.
Language:en_ZA
Published: 2014
Subjects:
Online Access:http://hdl.handle.net/10413/11416
Description
Summary:In this study I explore the extent of computer literacy in South Africa, the correlates of computer literacy, and the relationship between computer literacy and labour market outcomes, namely the probability of employment and earnings among working-age South Africans. I use data from the first wave of the National Income Dynamics Panel survey of 2008, the first national household survey to collect information on computer skills. This study focuses on computer literacy as it has become an integral skill in today’s world of fast technological change. Understanding the unequal distribution of this form of human capital and the benefits it affords those in the labour market, is important particularly in South Africa, where there is a growing gap between the rich and the poor. I find that the distribution of computer skills in South Africa tends to mirror existing inequalities; females, Africans, those with low levels of schooling and those living outside of formal urban areas, for instance, are less likely to be computer literate. Further, I find that there is a positive association between computer literacy and the probability of employment among working-age adults, and a positive relationship between computer literacy and earnings among the employed in South Africa. These associations hold after controlling for a variety of demographic, human capital, family background, and in the case of the earnings regressions, job characteristics. The results also suggest that, as would be expected, those who are highly computer literate do better than those who have basic use skills. I also consider the limitations of my methods and the data I use, and the implications of the results for education and skills development policy in South Africa. === M.A. University of KwaZulu-Natal, Durban 2013.