Logistic Regression Model and Decision Trees to Analyze Changes in Tourist Behavior: Tungurahua Case Study

Angélica González-Sánchez,Johanna Monge-Martínez,Leonardo Ballesteros-López,Sonia Armas-Arias
DOI: https://doi.org/10.1007/978-3-030-96046-9_16
2022-01-01
Abstract:This research was designed to identify the factors that influence the behavior changes of tourists visiting the province of Tungurahua and build the new visitor profile. The analytical method applied in this study was through the logistic regression model. Through the use of the CHAID logarithm, the decision trees were constructed. For this research, a total of 323 questionnaires were collected and validated by 4 experts. Additionally, the Cronbach’s Alpha statistic was used with a confidence level of 0.914. In this way, some questions were accepted for this research. Among the findings, it is identified that the behavior of tourists is determined by attitudes, decision conditions, travel destination choices, risk perceptions, and marketing factors. Decision tree models revealed that the choice of tourist destination depends on the effective application of biosafety regulations by service providers. Millennials are the potential market, coming from some cities of Ecuador, like Ambato, Quito, and Latacunga. They have a university or higher level of education and belong to a middle and upper-middle-income social stratum.
What problem does this paper attempt to address?