In this project, our data did not meet the assumptions of parametric tests, such as normality and equal variance of errors. Hence, we used non-parametric versions for our statistical tests. We have two dependent variables (comfort and usefulness) and one independent variable (FoV). To analyze our data, we conducted an overall Wilcoxon rank sum test (for two groups, between subjects), and Friedman rank sum test (for multiple groups, within subjects) across all conditions to analyze differences in the measured variables among the conditions. For the Friedman rank sum test, we performed pairwise Wilcoxon signed rank tests as the post hoc test using Benjamini-Hochberg (BH) correction for multiple comparisons.
The paper was published in ASSETS 2020.
Paper: [LINK]