Lessons for Asia from Mexico: understanding costumer satisfaction with fixed internet services by applying a factor analysis using polychoric correlations

The aim of this research consists on determine the factors that explain costumer satisfaction in the Mexican household fixed Internet service industry. The impossibility to measure directly our variable of interest entails the use of a factor analysis methodology which examines latent variables indirectly by employing measurable variables that are believed to be indicators of customer satisfaction. Data gathering is based on an anonymous questionnaire survey of 1481 adults interviewed by telephone conducted by a private consultant by request of the Mexican Federal Institute of Telecommunications on February 2016. The survey includes eight questions that gauge respondent's perceptions about the quality of services offered by his internet service provider. Subsequently, those opinions are coded on Likert scale of scores 1-5. These ordered-category data require the use of Polychoric correlation in order to find patterns between the indicators that allows us to infer their relationship to our unknown latent costumer satisfaction variable. Futher, the data gathered includes demographic characteristics of individuals and detailed information of internet service providers' characteristics, such as internet connection speed. The factor analysis was conducted with varimax rotation to obtain variable loads onto one factor as high as possible. The findings suggest the prevalence of two main factors: the first one related with technical features of the internet service and the second one with the service features provided by the internet service company.