The PISA2012 dataset comes from a triennial international survey conducted by Programme for International Student Assessment (PISA) aiming to evaluate education systems worldwide by testing the skills and knowledge of 15-year-old students from 65 economies.
In the exploration, I discovered that the correlation between out-of-school study hours and scores are pretty low. Family wealth seems to have a higher correlation with the average score compared with school learning hours. Surprisingly, the occupation status of parents has the highest correlation coefficient among all numeric variables. Students from most countries perform similarly, except China scores rather higher on average, and Brazil much lower on average. However, interestingly, China also has some lowest scoring students, which shows a large disparity between students in China. Gender doesn't seem to make any big difference to the average scores. On average, they perform similarly, except males can have a lower lowest scores while female can have a higher highest scores, but those are very rare. Family wealth and parents occupation status also has a strong correlation, which makes sense. Interestingly, family wealth and school learning time also seem to have a positive correlation. The study hours in school and out of school for Chinese and Japanese students don't stand out among others. Comparatively, China seem to have longer study hours out of school, but Russia and Italy also seem pretty long. Surpringly, Japan actually has rather short study hours even though they perform also pretty well. But it could also be that there are not as many participants from China and Japan, that's why we don't see a huge difference. For China, US, and Brazil, there seems to some positive correlation between wealth and average scores. It makes perfect sense, since these are the countries that have very large economic inequality, poorer kids are probably less accessible to learning resources. There isn't any country that stands out when it comes to parent occupation status and scores, but the overall correlation is positive. Finally, there is also not much difference between genders on how school study hours, wealth, and parents status affect scores.
The average scores seem to be most strongly correlated with the parents occupation status and then family wealth. The study hours be it at school or out of school have a rather weak correlation. As parents occupation status and family wealth increase, scores tend to increase as well. Surprisingly, study hours are not a key factor in determining the average scores. Even in some countries where study hours are shorter, the scores are higher, and the opposite scenario also happens. From my analysis, students from better family background do tend to have higher scores especially in countries where economic disparity is large.