An investigation of affect of some variables on the science score achievement by quantile regression based on PISA 2015 Turkey data

In this master thesis study, it was investigated that whether the affect of some variables (EPIST, EMOSUPS, UNFAIRTEACHER, DISCLISCI, HISCED, HISEI, CULTPOSS, HEDRES, ESCS) on science literacy score by using PISA 2015 Turkey data according to the different regions (Istanbul, Mediterranean and Southeastern Anatolia).Furthermore, it was investigated whether the effect of these variables was different in low, medium and high score students. The study was carried out using data from 4486, 852, 636 and 453 students for the Turkey, Istanbul, Mediterranean and Southeastern Anatolian regions, respectively, after eliminating the missing data from the PISA 2015 Turkey application data. As a result of the analysis, it was observed that the effect of HISCED, HISEI, CULTPOSS, HEDRES, ESCS and EPIST variables was differentiated from other regions in the Mediterranean region. While the effects of HISCED and ESCS variables were higher in the Mediterranean region compared to the other regions, HISEI, CULTPOSS, HEDRES and EPIST variables were observed to be less affected than the other regions. EMOSUPS and UNFAIRTEACHER variables were observed to be higher in Southeast Anatolia than in other regions. It is observed that the effect of DISCLISCI variable in all three regions is higher than Turkey average. It was observed that the effect of the ESCS variable did not differ much in all regions in the students with low, medium and high scores. It was observed that the effect of the EPIST variable was obviously less in the students who had high scores in the Istanbul region than those who had medium and low scores. It was also observed that the effect of this variable was less in the Mediterranean region than in the medium and high-rated students. While the effect of the EMOSUPS variable does not show a significant change in Turkey's general, Istanbul and Mediterranean regions in the students with low, medium and high scores, it is observed that the students in the Southeast Anatolia region have a very high effect on the students with a medium score. It is observed that the UNFAIRTEACHER variable has a lower impact on middle and high performing students in the Southeastern Anatolia region than in the other regions. It is observed that the HISCED variable has more impact on middle and lower students in the Mediterranean region than in the other regions. Likewise, the CULTPOSS variable was observed to be more affected in the median and highscore students in the Mediterranean region than in the other regions. The negative and effect of the HEDRES variable is high in middle and high scores in the Mediterranean region, whereas it is observed that there is a positive coefficient in the students in high scores in Southeast Anatolia.