This paper examines the PISA 2006 data for Greece. The paper focuses on the effect of specific socio-economic characteristics in the Science performance. The research question relies on the potential relation that the above effect will bring out. Linear regression models were used to assess the effect of the aforementioned socio-economic parameters on students’ performance after age and gender adjustment standardised beta coefficients were calculated and assisted in hierarchy the effect of these factors (the higher, the more effective). For this purpose three indices were developed (an index indicating the family’s wealth status and two indices to evaluate students’ use of computer for educational or entertaining purposes). Parents’ educational status and the number of books were also considered. Results of linear regression revealed that students’ performance increases as the number of books increases, as father’s and mother’s educational status improves and as the wealth status of a family is improving; while a frequent use of computer for educating purposes decreases the mean science performance. In this paper, the theory considering knowledge as socially determined is arisen as well as the view that the capital (of socio-economic features) that is ‘imparted, increased, decreased or disappears’, affects knowledge construction and production.