ESTG - Mestrado em Ciência de Dados
Permanent URI for this collection
Browse
Browsing ESTG - Mestrado em Ciência de Dados by advisor "Ferreira, Susana Raquel Carvalho"
Now showing 1 - 1 of 1
Results Per Page
Sort Options
- Characterization of the lifestyle and well-being of students from the Polytechnic of LeiriaPublication . Santos, Daniel Augusto Bertoldo; Santos, Rui Filipe Vargas de Sousa; Ferreira, Susana Raquel CarvalhoEntering higher education marks a significant juncture in a student’s life. It often involves a shift towards independence, characterized by a distancing from family and friends, increased responsibilities, and greater autonomy in decision-making. These changes can influence well-being and various aspects of lifestyle, such as dietary habits, exercise routines, alcohol and drug use, and sexual behavior. Despite the initial excitement, this transition may also induce stress and anxiety. Academic demands, including grades, exams, and deadlines, as well as the newfound responsibilities of managing one’s schedule, finances, and social relationships, all affect the well-being. Hence, several studies recently conducted on college students have highlighted the importance of monitoring their well-being, especially since several reports have indicated a significant increase in mental health issues among college students, such as depression and anxiety. In particular, the Short Multidimensional Inventory Lifestyle Evaluation (SMILE), developed in 2020, is a 43-item self-rated questionnaire consisting of 7 domains, allowing a multidimensional evaluation of a (healthy) lifestyle. Within this context, a web survey was conducted among students at the Polytechnic of Leiria. This survey collected socio-demographic data, SMILE scores and clinical variables data, including screening for depression and anxiety. The key insights gleaned from the statistical analysis of the obtained data are summarized, particularly focusing on describing lifestyle and well-being, discerning differences between categories, and validating the survey instrument. Two supervised learning classification methodologies (logistic regression and decision trees) were applied to identify depression and anxiety issues based on responses to the survey. The reliability of these classifications were carry out using confusion matrix, accuracy, sensitivity, specificity, predictive values, and the area under the ROC curve in a test sample. The results reveal that lower SMILE scores are associated with positive screening of depression/anxiety in higher education students, despite the reliability appears insufficient to confidently recommend its use for screening depression and/or anxiety disorders. However, it enables the characterization of students’ lifestyles, the assessment of their well-being levels, and, consequently, the identification of potential mental health issues.