Modelo de mensuração do comportamento empreendedor a partir das características comportamentais e intenção empreendedoras

Abstract

The traditional methods of evaluating entrepreneurial behavior carry a degree of uncertainty and significant subjectivity with several uncontrollable independent variables. In face of this fragility, the fuzzy logic can be a tool that contributes to the understanding of the degree of pertinence regarding the entrepreneurial behavior of the individual. In order to contribute in this direction, this study aimed to develop a model of measurement of the entrepreneurial behavior from behavioral characteristics and entrepreneurial intention. An applied, quantitative, descriptive and exploratory study was carried out. In order to satisfy the problem raised, already validated data collection instruments were used, the questionnaire on entrepreneurial behavioral characteristics (EBC), developed by McClelland (MANSFIELD et al., 1987), and the questionnaire of entrepreneurial intentions of Liñán and Chen (2009). These instruments were answered by undergraduate students of the Federal University of Santa Maria. The sample consisted of 2,519 research subjects. The internal consistency of the scales was measured by the Alpha Cronbach Coefficient, ensuring reliability to the measurement and evaluation of the constructs. Initially the analysis was performed through descriptive statistics for entrepreneurial behavioral characteristics and for entrepreneurial intention the traditional method, where the behavioral characteristics and respective dimensions of the students were identified. It was verified that more than 90% of the students present entrepreneurial behavioral characteristics and that they intend to start a business. Then, through the modeling of structural equations, the model for the measurement of entrepreneurial intention was revalidated. From this, behavioral characteristics and entrepreneurial intention were related and a structural model for entrepreneurial behavior was developed. Based on the modeling of structural equations for entrepreneurial behavior, the premises for the development of the fuzzy model were elaborated. In this study, a fuzzy model was developed to measure entrepreneurial behavior through entrepreneurial behavioral characteristics and entrepreneurial intention. The model developed follows a set of linguistic variables, pertinence functions, fuzzification method and defuzzification, which are the component elements of the fuzzy logic. The fuzzy modeling was built in MATLAB from the Fuzzy Toolbox. For the construction of this fuzzy model of measurement of the entrepreneurial behavior the Mamdani method was used, with defuzzification by means of the Centroid method. The fuzzy model of measurement of entrepreneurial behavior was configured and then tested in Simulink environment. In fuzzy modeling, the percentage of students with entrepreneurial behavior fell to 83%; in addition, 4% were considered "perhaps entrepreneurs" and 13% of students did not present entrepreneurial behavior. The fuzzy model of measurement of entrepreneurial behavior proved adequate for such measurement. Finally, the results of the fuzzy measurement model developed and the traditional method of evaluation by the average were compared, where it was verified that the fuzzy modeling is more detailed and more reliable compared to the traditional analysis method. It is concluded that the objective was reached and that the fuzzy model developed is adequate to measure entrepreneurial behavior. In an unprecedented way, the fuzzy model of measurement of the entrepreneurial behavior can contribute as an alternative tool in the behavioral evaluation of the individuals. From this research, the search for different ways of analyzing the entrepreneurial behavior against the traditional statistical models used is promising. The research was limited to the construction of a model of measurement of entrepreneurial behavior from the instruments answered by undergraduate students from a public higher education institution. Future studies may expand the constructs addressed in the fuzzy model, as well as be replicated in other institutions of higher or basic education in order to compare, deepen and make new findings, in addition to considering a longitudinal follow-up of the entrepreneurial behavioral evolution of the individuals surveyed.