Examiination of the predictive powers of socio-cultural and socio-economic variables for PISA science literacy by years

The purpose of this research is to determine whether there are variables consistntly predicting science literacy in the scope of 15-year-old Turkish students' socio-cultural and socio-economic characteristics (parental education, positions of business, cultural possessions, home possessions) according to administration period of PISA (2003, 2006, 2009, 2012) or not; and if any to determine these variables. The population of this research – designed in the cross-sectional approach of time series patterns-consists all Turkish students at the age range of 15 years (plus 3 months) old to 16 years (plus 2 months) old. The sample of the research consists of 4855 selected students for PISA 2003 administration, 4942 selected students for PISA 2006 administration and 4996 selected students for PISA 2009, 4848 selected students for PISA 2012 administration through stratified sampling method. The variables predicting Turkish students' PISA science literacy scores were determined by using common items taking place in 2003, 2006,2009 and 2012 student questionnaire forms designed to examine socio-cultural characteristics. Data of this research were analyzed with stepwise multiple regression analysis via Statistical Package for the Social Sciences (SPSS) 20.0. In conclusion, it is found that socio-cultural and socio-ekonomic variables being analyzed in this study are significant predictors for all PISA administrations; and all of variables including parental education, positions of business, cultural possessions and home possessions are found as consistent predictors for PISA science literacy scores of Turkish students consistently in 2003, 2006, 2009 and 2012 administtation periods.