Factors influencing university drop out rates.
ABSTRACT This paper develops personalized models for different university degrees to obtain the risk of each student abandoning his degree and analyzes the profile for undergraduates that abandon the degree. In this study three faculties located in Granada, South of Spain, were involved. In Software Engineering three university degrees with 10,844 students, in humanities nineteen university degrees with 39,241 students and in Economic Sciences five university degrees with 25,745 students were considered. Data, corresponding to the period 1992 onwards, are used to obtain a model of logistic regression for each faculty which represents them satisfactorily. These models and the framework data show that certain variables appear repeatedly in the explanation of the drop out in all of the faculties. These variables are, among others, start age, the father’s and mother’s studies, academic performance, success, average mark in the degree and the access form and in some cases also, the number of rounds needed to pass. Students with weak educational strategies and without persistence to achieve their aims in life have low academic performance and low success rates and this implies a high risk of abandoning the degree. The results suggest that each university centre could consider similar models to elaborate a particular action plan to help lower the drop out rate reducing costs and efforts. As concluded in this paper, the profile of the students who tend to abandon their studies is dependent on the subject studied. For this reason, a general methodology based on a Data Warehouse architecture is proposed. This architecture does most of the work automatically and is general enough to be used at any university centre because it only takes into account the usual data the students provide when registered in a course and their grades throughout the years.
- SourceAvailable from: Cristóbal Romero[show abstract] [hide abstract]
ABSTRACT: On-line discussion forums constitute communities of people learning from each other, which not only inform the students about their peers' doubts and problems but can also inform instructors about their students' knowledge of the course contents. In fact, nowadays there is increasing interest in the use of discussion forums as an indicator of student performance. In this respect, this paper proposes the use of different data mining approaches for improving prediction of students' final performance starting from participation indicators in both quantitative, qualitative and social network forums. Our objective is to determine how the selection of instances and attributes, the use of different classification algorithms and the date when data is gathered affect the accuracy and comprehensibility of the prediction. A new Moodle's module for gathering forum indicators was developed and different executions were carried out using real data from 114 university students during a first-year course in computer science. A representative set of traditional classification algorithms have been used and compared versus classifi-cation via clustering algorithms for predicting whether students will pass or fail the course on the basis of data about their forum usage. The results obtained indicate the suitability of performing both a final prediction at the end of the course and an early prediction before the end of the course; of applying clustering plus class association rules mining instead of traditional classification for obtaining highly interpretable student performance models; and of using a subset of attributes instead of all available attributes, and not all forum messages but only students' messages with content related to the subject of the course for improving classification accuracy.Computer&Education. 01/2013; 68:458-472.
- [show abstract] [hide abstract]
ABSTRACT: The purpose of this study is to examine at-risk students and the reasons they give to explain their poor academic performance, with a view to developing a typology of at-risk students. A case study methodology was used to investigate the total population of at-risk students for Semester 2, 2008 studying at the Singapore campus of an Australian-based university. Poor academic performance means that students are placed ‘at-risk’ of exclusion from the University if their grades do not significantly improve in subsequent semesters. The majority of students cite employment pressures (primarily work commitments interfering with study) and personal relationship difficulties (including divorce and family commitments) as the main causes of their at-risk status. Our findings may help universities implementing at-risk programmes reduce student attrition and better aid students in completing their degrees.Journal of Higher Education Policy and Management 02/2012; 34(1):3-13.
- [show abstract] [hide abstract]
ABSTRACT: Predicting student failure at school has become a difficult challenge due to both the high number of factors that can affect the low performance of students and the imbalanced nature of these types of datasets. In this paper, a genetic programming algorithm and different data mining approaches are proposed for solving these problems using real data about 670 high school students from Zacatecas, Mexico. Firstly, we select the best attributes in order to resolve the problem of high dimensionality. Then, rebalancing of data and cost sensitive classification have been applied in order to resolve the problem of classifying imbalanced data. We also propose to use a genetic programming model versus different white box techniques in order to obtain both more comprehensible and accuracy classification rules. The outcomes of each approach are shown and compared in order to select the best to improve classification accuracy, specifically with regard to which students might fail.Applied Intelligence 08/2013; 38(3):315-330. · 1.85 Impact Factor