An association model for bivariate data with application to the anlysis of university students' success.

Research output: Contribution to journalArticlepeer-review

4 Citations (Scopus)

Abstract

The academic success of students is a priority for all universities. We analyze the students' success at university by considering the performance in terms of both 'qualitative performance', measured by their mean grade, and 'quantitative performance', measured by university credits accumulated. These data come from an Italian university and concern a cohort of students enrolled at the Faculty of Economics. To jointly model both the marginal relationships and the association structure with covariates, we fit a bivariate ordered logistic model by penalized maximum likelihood estimation. The penalty term we use allows us to smooth the association structure and enlarge the range of possible parametrizations beyond that provided by the usual Dale model. The advantages of our approach are also in terms of parsimony and parameter interpretation, while preserving the goodness of fit.
Original languageEnglish
Pages (from-to)46-57
Number of pages12
JournalJournal of Applied Statistics
Volume43
Publication statusPublished - 2016

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Fingerprint Dive into the research topics of 'An association model for bivariate data with application to the anlysis of university students' success.'. Together they form a unique fingerprint.

Cite this