Appropriate statistical tests for comparative hypotheses in Plant Science Research

Choosing the appropriate statistical methods is one of the most important steps in designing a quantitative research method. Although ANOVA is one of the most widely used statistical analyses in plant ecology, violations of its assumptions can cause various issues, like statistical errors and biased estimates. The present research was carried out in two experiments in the Artemisia habitat in Sarbisheh. In the first experiment, to investigate grazing intensity effect on vegetation cover , production, species density and richness, ANOVA, Kruskal-Wallis, Welch, Brunner-Dette-Munk, and one-way permutation tests were used, respectively. In the second experiment, in order to investigate the effect of the range management plan and grazing intensity on the vegetation cover, the two-way ANOVA test in the factorial and split plot were used, and for species density, the two-way BDM test and non-parametric tests for general factorial designs were used. The results showed that the effect of grazing intensity on all vegetation characteristics was significant. The greatest effect of grazing intensity was on production . The highest vegetation cover was observed in light grazing . The results showed that the effect of the interaction on the vegetation cover and species density were not significant. In the present study, instead of , Welch and non-parametric tests were used. The results of most of the tests were similar to the results of ANOVA. In situations where it is not possible to use parametric statistical tests, it is recommended to remove outlier data, transform data, use robust tests and use non-parametric methods.