Modelling the assessment of taxpayer perception on the fiscal system by a hybrid approach for the analysis of challenging data structures

Models of tax compliance analysed various factors affecting tax compliant behaviour, from human internal motivations to public perception, risk aversion of penalty and trust in State. For tackling the assessment of taxpayer perception on the fiscal system in a challenging survey based on multiple items, a hybrid statistical model is introduced. In particular, after verifying for each individual the component of feeling and uncertainty in the response process to individual blocks of items intended to measure common latent traits, we synthesized the information in meta-items applying a data-reduction. The meta-items are then modelled according to statistical frameworks for ordered polytomous variables accounting for potential uncertainty in the process of response. Our results—based on a sample of 366 respondent-students attending various Finance courses in Europe—display a certain gender bias in uncertainty levels showing that women feel more uncertain when expressing their opinion on the taxation system. Furthermore they are in accordance with previous studies showing that the trust in State supports voluntary tax compliance and it is driven by clearer laws, transparent communication, perceived good quality of public services and efficient policies for ensuring social welfare of citizens.