Identifying Determinant Factors for Students’ Success in Preparatory Schools Using Data Mining Techniques
No Thumbnail Available
Date
2017-06
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
A.A.U
Abstract
Data from Educational Assessment and Examination Agency shows that from Addis Ababa region who took Ethiopian Higher Education Entrance examination in 2006, 2007 and 2008 E.C; 64.1 %, 54.4 %, and 42.4% respectively scores less than half of the total score. Even though the percentage decreases from 2006 E.C to 2008 E.C.by 21.7%, many students fail to score the expected half of the total score.
The study aims to apply data mining for identifying the determinant factors for the students’ success in the preparatory schools to join higher education. The study focused on Addis Ababa region Natural Science stream preparatory Schools’ students. Based on this, the data collected from National Educational Assessment and Examination Agency is only Addis Ababa examinees EHEEE data and correspondingly their EGSECE data. The collected dataset covers three years of data from the three years data from 2006 to 2008 E.C EHEEE.
The study uses Hybrid data mining model since it is a research oriented model and WEKA 3.8.0, Microsoft Excel 2013 and KU tools are used for data mining, for data integration and for data exploration respectively. Finally, 40328 instances and 15 attributes are selected for analysis. Additionally, the values of some of the attributes are discretized using the assessment system of secondary education in Ethiopia which is categorized as Excellent, Very good, Good, Satisfactory and Fail. Association rule mining method such as Apriori and Filter Associator algorithm compared and Apriori algorithm is applied in order to get the results. By configuring different thresholds, different rules are achieved. The discovered rules are then evaluated using the interestingness measure lift or correlation and domain experts.
As a result, the study showed that scoring Very good in Physics, Civics and Biology subjects in EHEEE are determinant factors for the students ‘success in the preparatory schools. Similarly scoring good in English in EHEEE is also another determinant factor. Besides, the study revealed that Regular Non-Government preparatory students are more associated with success to enter higher education than Government preparatory school students and Sub city of schools that students attend have no influence on students’ success to enter higher education.
But in using Apriori algorithm, there is no standard way of setting different thresholds. This leads to missing the strong rules.
Description
A Thesis Submitted to the School of Graduate Studies of Addis Ababa University in Partial Fulfillment of
The Requirements for the Degree of Master
Of Science in Information Science
Keywords
Data, Data Mining Techniques