Use of Data Mining to Determine Usage Patterns of an Online Evaluation Platform During the COVID-19 Pandemic

MenPas is a psychosocial assessment platform developed by the University of Malaga in 2008. There has been a significant increase in data traffic during the period of confinement by COVID-19 (March and April '20) compared to the same period in the previous year. The main goal to achieve in this work is to determine the patterns of use of this platform on both period of time. So, we want to respond to the following question: So, we the following question: Has the COVID-19 Pandemic changed the pattern of the Menpas users? In order to respond it, cluster analysis techniques (Data Mining) have been used to classify people taking surveys into quotient sets (cluster). This is a multivariate technique for dividing data into sets to that are as homogeneous as possible within themselves and heterogeneous among themselves. Specifically, the K-Means algorithm has been used for this analysis, which is based on the evaluation of the distance between data and the average of each variable. So, it is recommended to discover patterns or relationships among the data. Specifically, the use of the following questionnaires has been analyzed: Competitive State Anxiety Inventory-2 (CSAI-2), State Trait Anxiety Inventory (STAI), Profile of Mood State (POMS), Resilience Scale (RS), Sport Performance Psychological Inventory (IPED), Maslach Burnout Inventory (MBI) and Self-concept Form-5 (AF-5). The analyses have shown changes in cluster formation between 2019 and 2020 based on the variables gender, age, marital status or physical practice. Therefore, the analyses carried out have been sensitive to determine several profiles of people using the MenPas platform because there are changes in the characteristics of the user groups that have carried out the analyzed tests.