A data set of the attributes of 382 students in secondary education collected from two schools. The goal is to predict the grade in math and portugese at the end of the third period. See the cited sources for additional information.
student
382 observations from 13 variables represented as a list consisting
of a binary factor response matrix y with two responses: portugese and
math for the final scores in period three for the respective subjects.
The list also contains x: a sparse feature matrix of class
'dgCMatrix' with the following variables:
student's primary school, 1 for Mousinho da Silveira and 0 for Gabriel Pereira
sex of student, 1 for male
age of student
urban (1) or rural (0) home address
whether the family size is larger than 3
whether parents live together
mother's level of education (ordered)
fathers's level of education (ordered)
whether the mother was employed in health care
whether the mother was employed as something other than the specified job roles
whether the mother was employed in the service sector
whether the mother was employed as a teacher
whether the father was employed in health care
whether the father was employed as something other than the specified job roles
whether the father was employed in the service sector
whether the father was employed as a teacher
school chosen for being close to home
school chosen for another reason
school chosen for its reputation
whether the student attended nursery school
Pwhether the student has internet access at home
P. Cortez and A. Silva. Using Data Mining to Predict Secondary School Student Performance. In A. Brito and J. Teixeira Eds., Proceedings of 5th FUture BUsiness TEChnology Conference (FUBUTEC 2008) pp. 5-12, Porto, Portugal, April, 2008, EUROSIS, ISBN 978-9077381-39-7. http://www3.dsi.uminho.pt/pcortez/student.pdf
Dua, D. and Karra Taniskidou, E. (2017). UCI Machine Learning Repository http://archive.ics.uci.edu/ml. Irvine, CA: University of California, School of Information and Computer Science.
All of the grade-specific predictors were droppped from the data set. (Note that it is not clear from the source why some of these predictors are specific to each grade, such as which parent is the student's guardian.) The categorical variables were dummy-coded. Only the final grades (G3) were kept as dependent variables, whilst the first and second period grades were dropped.