The paper explores the role of human and non-human capital factors in determining the earnings of workers in Pakistan. The Labour Force Survey data 1993-94, which provides detailed information about workers, has been used for this analysis. Ordinary least squares estimation technique has been utilised for the analysis. The analysis is carried out separately for male and female workers. It is found that in human capital variables, education plays a dominant role in wage determination. In non-human capital factors, occupation and size of the establishments are found to be relevant variables. Other important variables include regional location and technical training. The sample selection bias has been observed for female workers whereas no problem has been observed for male workers, and the bias has been corrected by using the Heckman procedure, though both corrected and uncorrected results are reported. Important conclusions and policy implications are discussed at the end.