Response variable: overall_score
Scenario
It is a difficult task to find an overall measure of quality for higher-education institutions, as there are many areas of work most universities worldwide involve themselves in, such as teaching, research, and knowledge exchange. Nevertheless, to decide on a way to predict the overall quality of an institution would be desirable for those who want to make an informed decision of whether to engage with a specific university, as a student, research collaborator or as industry partner.
This dataset contains records of many universities, offering data on performance metrics, student data and descriptive data. The goal is to forecast the value for the variable overall_score, by employing models suitable for regression problems. Find out what characteristics have the most influence on the general quality of institutions of higher learning.
Columns Description
name: Name of the university
scores_teaching: The teaching quality of the university, scored out of 100
scores_research: The research quality of the university, scored out of 100
scores_citations: Citation volume and of academics based at the university, scored out of 100
scores_international_outlok: The university’s level of engagement with international partners, scored out of 100
record_type: Category of the record
member_level: Level of membership
location: Country where the university is located
stats_number_students: The number of students enrolled at the university
stats_student_staff_ratio: Number of staff members per student
stats_pc_intl_students: The percentage of enrolled students that are classed as international
stats_female_male_ratio: The ration of female students versus male students
subjects_offered: The range of subject areas that are taught at the university
closed: Whether the university is currently closed to new
applicants
unaccredited: Whether the university is currently unaccredited
overall_score: The overall quality of the university, scored out of
100