Do university students really need to be taught by the best instructors to learn?

Kalender, İlker

doi:10.1080/2331186x.2017.1389334

Cited by 1 publication

(1 citation statement)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The CHAID algorithm has specific advantages over statistical inferential analysis such as it does not need to fulfil several assumptions (Baran & Kihç, 2015), for e.g. normality, linearity, additivity and homogeneity (Karakaya, Mehmet, Corbaci, & Cetin, 2018), it automatically select predictor variables that could be used to define a subgroups, detects nonlinear association (Kalender, 2017) and the detection includes the association between categorical dependent variable and multiple independent variables which can be categorical and/or metric (Milanović & Stamenković, 2016). Decision tree analysis especially CHAID have been found to be the most robust against changes in data accuracy within the context of customer market segmentation (Coussement, Van den Bossche, & De Bock, 2014), or in this study university students' spending behavior.…”

Section: Literature Reviewmentioning

confidence: 99%

Spending profiles of sultan idris education university undergraduates: a CHAID-based segmentation

Khalid

Ismail

2019

IBEJ

View full text Add to dashboard Cite

Issues on college students spending behaviour have been studied numerous times due to improper use of education funds. Various studies examined determinant factors for this misbehaviour in order to understand, identify and recommend ways to educate these future generations. Some documented relationships between financial behaviours and financial literacy, money beliefs and socialization agents. This study walks the same path but focuses on spending behaviour rather than financial behaviour and examines its association with determinant factors by means segmentation analysis. The aim of the study is to profile Sultan Idris Education University undergraduates' according to their spending using a decision tree analysis procedure called Chisquared Automatic Interaction Detector (CHAID). CHAID generates predictive tree models by segmenting based on predictor variables which include demographic characteristics, program of study related traits and funding aspects that were utilized to profiles students' spending. Spending in this study excludes most incurred expenditures namely living and subsistence costs. Spending profiles were determined based on 751 feedbacks received from business students of Sultan Idris Education University. It was found that, top five most incurred non-subsistence expenses were personal hygiene products, study materials, telecommunications services, clothing and health products. Interestingly, all these five types of spending can be segmented according to only one predictor variable namely, faculty. It can be assumed from this results that, spending profiles are shared within the same faculty. Since spending profiles can be differentiated between faculties, it can be recommended that education programs directed at influencing spending behaviour should be tailor-made according to faculties.

show abstract

Section: Literature Reviewmentioning

confidence: 99%