• Login
  • Register

Work for a Member company and need a Member Portal account? Register here with your company email address.

Project

Mapping Higher Education Option Space

The mismatch between the supply of graduates’ skills and the needs of the labor market has become increasingly obvious and problematic over the past years. This has prompt policymakers, educators, employers and applicants to reevaluate the role of higher education system. But how do each of these actors perceive higher education? How similar are, according to them, the different degree programs and institutions?

In this project, we use a data-driven approach to unveil the structure of similarities between degree programs as perceived from the candidates. To that end, we use applicants’ preferences to higher education in Chile and Portugal between the years of 2007 and 2014 as a proxy to measure the similarity between each pair of degree programs. We find that:

  1. The two structures share the same topological features, despite coming from two different political and social-economical contexts;
  2. We quantify the mismatch between the current state of the art classification used by educators and policymakers and the structure identify
  3. We find the existence of strong spatial patterns in the assortment of gender, application scores, demand and unemployment levels; and
  4. We find that structure of similarities encapsulates non-trivial information about the nature of each degree program, allowing us to predict with high accuracy the level of unemployment by just taking into account the relative position of a degree program in the higher education option set. 

Currently, we are preparing a manuscript to present our findings. 

Copyright

CC

In the figure, we can see the networks of relationships between majors in the Portuguese Higher Education System between 2008 and 2015, and Chilean Higher Education System between 2007 and 2015.
A) Graphical representation of the network for Portugal, nodes are colored according to the ISCED classification while their sizes are proportional to the degree.
B) Graphical representation of the network for Chile, nodes are colored according to the ISCED classification while their sizes are proportional to the degree.
C) The spatial correlation for gender dominance, average scores of applicants, supply over demand level and unemployment level for Portuguese system. The figures show the level of correlation on the y-axis and the degrees of separation on the x-axis.
D) The spatial correlation for gender dominance, average scores of applicants, and supply over demand level for Chilean system. The figures show the level of correlation on the y-axis and the degrees of separation on the x-axis. (We don't have information about unemployment in Chile.)

Research Topics
#data #networks