ERIC Number: ED490086
Record Type: Non-Journal
Publication Date: 2004-Apr-10
Identifying Characteristics of High School Dropouts: Data Mining with A Decision Tree Model
Veitch, William Robert.
Online Submission, Paper presented at the Annual Meeting of the American Educational Research Association (62nd, San Diego, CA, April 10-14, 2004)
The notion that all students should finish high school has grown throughout the last century and continues to be an important goal for all educational levels in this new century. Non-completion has been related to all sorts of social, financial, and psychological issues. Many studies have attempted to put together a process that will identify students at risk of dropping out by using various research methodologies. The purpose of this study is to investigate correlates of high school dropping out through the use of data mining of existing data sources with decision trees. Decision tree methods are designed to sift through a set of predictor variables and successively split a data set into subgroups in order to improve the prediction (classification) of a target (dependent) variable. As such, these methods are valuable to data miners faced with constructing predictive models when there may be a large number of predictor variables and not much theory or previous work to guide them. The tree presented in this paper does exhibit a certain ability to predict which students may drop out of school.
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: High Schools
Authoring Institution: N/A
Grant or Contract Numbers: N/A