Metodologia testu chi-kwadrat na przykładzie badań ankietowych dotyczących europejskich serów regionalnych

Abstract

The aim of the research was to present the methodology of the chi-square test of independence with the use of the Statistica package on the example of the results of a survey on European regional cheeses. Three groups of data were obtained: 10 qualitative (categorizing) variables, 4 variables as multiple responses, 46 variables as multiple dichotomies, 60 variables in total. Each questionnaire answer was assigned 4 differentiators describing the respondent, i.e. gender, age, education, and his place of residence. The Pearson chi-square test and the maximum likelihood chi-square (NW) test for a total of 240 duplicate tests were analyzed and non-parametric Spearman rank correlations between the selected variables were calculated. On the basis of the obtained results, the existence of 25 statistically significant relationships between the analyzed variables and one of the four characteristics of the respondents' record was proved. The analysis of Spearman's rank correlation between the selected variables showed at most a moderate correlation between several pairs of variables. The obtained results confirmed the usefulness of the chi-square test of independence and the analysis of Spearman's rank correlation in the survey.(original abstract)