Application of the CART Model to Classify the Perception of Young Canadian Teenagers on the Effect of Marijuana on Health

Authors

  • Ruben Thoplan Department of Economics and Statistics, Faculty of Social Studies and Humanities, University of Mauritius, R

Keywords:

marijuana, CART, pruned, psilocybin, ROC curve

Abstract

The use of the illicit drug, marijuana has increased over years among young teenagers in different parts of the world and its harm on the health is generally well-known. This paper attempts to study the perception of young adolescents of 13-15 years old residing in Canada towards the danger of marijuana on health. To do so, a classification and regression tree (CART) has been applied on the data from the 2012 National Anti-Drug Strategy (NADS) Youth Advertising Recall and Tracking Survey. The decision tree has been applied and pruned on a training data set (70%) and evaluated on the testing data set (30%). The results show that the main indicators which impact on the perception of a teenager towards the harm marijuana has on health are the perceptions towards psilocybin (another illicit drug), the province in which the teenager lives and whether he/she has been ever offered drugs. The overall error rate on the testing data set based on the confusion matrix is less than 20% and the area under the ROC curve is relatively high showing that the model is accurate in classifying the perception of young teenagers on the health marijuana has on health.


Author Biography

Ruben Thoplan, Department of Economics and Statistics, Faculty of Social Studies and Humanities, University of Mauritius, R

Ruben Thoplan is a lecturer in Statistics at the University of Mauritius since year 2009. He is currently research active in the field of data mining.

References

D. M. Fergusson, L. J. Horwoord and N. Swain-Campbell,

Unicef Office of Research,

Government of Canada,

Government of Canada,

Government of Canada,

NSDUH,

G. Williams, Data Mining with Rattle and R: The Art of Excavating Data for Knowledge Discovery, Australian Capital Territory: Springer, 2011.

K. R. Hess, M. C. Abbruzzese, R. Lenzi, M. N. Raber and J. L. Abbruzzese,

T.-S. Lee, C.-C. Chiu, Y.-C. Chou and C.-J. Lu,

G. De'ath and K. E. Fabricius,

L. Breiman, J. Friedman, C. J. Stone and R. A. Olshen, Classification and Regression Trees, Chapman and Hall/CRC, 1984.

T. M. Therneau and E. J. Atkinson,

R Development Core Team (2008). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org.

Downloads

Published

2014-08-09

How to Cite

Thoplan, R. (2014). Application of the CART Model to Classify the Perception of Young Canadian Teenagers on the Effect of Marijuana on Health. International Journal of Sciences: Basic and Applied Research (IJSBAR), 17(2), 236–242. Retrieved from https://gssrr.org/index.php/JournalOfBasicAndApplied/article/view/2648

Issue

Section

Articles