Validation and dimensional analysis of the eating behaviour pattern questionnaire among Malaysian university students

@#Introduction: Eating behaviour pattern is among the key behavioural factors that contribute to eating disorders. Hence, to evaluate the psychometric characteristics of the Eating Behaviour Pattern Questionnaire (EBPQ) that is used in epidemiological studies to measure the relationship between health outcomes and eating behaviour patterns, this study aimed to validate the adopted version of the EBPQ and to check the validity and reliability of this tool in University of Malaya, Malaysia. Methods: Exploratory factor analysis (EFA) was used to determine the most appropriate factor structure of EBPQ. Moreover, structural equation modelling (SEM) and confirmatory factor analysis (CFA) were applied to examine the convergent and discriminant validity of EBPQ. As for the participants of the study, multi-stage random sampling was used and 200 students (109 females and 91 males) from University of Malaya were chosen. Results: The EFA yielded nine components of EBPQ including emotional eating, eating outside, cultural habit, low-fat eating, meal skipping, snacking, healthy eating, planning for food and sweets, which explained 67.7% of the total variance. Furthermore, the Cronbach’s α was about 0.8 for all components, which exhibited a high internal consistency among the obtained components. The results showed that the questionnaire had sufficient convergent and discriminant validity. Conclusion: The EBPQ was proven to be a reliable tool to measure the eating behaviour patterns in Malaysian university students. The presence of adequate validity and reliability supports this instrument’s psychometric properties for future studies.

Feature selection methods for optimizing clinicopathologic input variables in oral cancer prognosis.

The incidence of oral cancer is high for those of Indian ethnic origin in Malaysia. Various clinical and pathological data are usually used in oral cancer prognosis. However, due to time, cost and tissue limitations, the number of prognosis variables need to be reduced. In this research, we demonstrated the use of feature selection methods to select a subset of variables that is highly predictive of oral cancer prognosis. The objective is to reduce the number of input variables, thus to identify the key clinicopathologic (input) variables of oral cancer prognosis based on the data collected in the Malaysian scenario. Two feature selection methods, genetic algorithm (wrapper approach) and Pearson's correlation coefficient (filter approach) were implemented and compared with single-input models and a full-input model. The results showed that the reduced models with feature selection method are able to produce more accurate prognosis results than the full-input model and single-input model, with the Pearson's correlation coefficient achieving the most promising results.