hi Soumyajit,

1.) parametric tests are not based on signs or ranks. you must have meant 'non-parametric' methods. Nonparametric test make less strict demands of the data. The central limit theorem does not apply strictly. When we apply parametric tests, underlying conditions or assumptions must be met, particularly for smaller sample sizes. For example, for the one-sample t test (comparing a sample mean to make an inference about a population parameter such as the population mean), for example, requires that the observations be drawn from a normally distributed population. For two independent samples, the t test has the additional requirement that the population standard deviations be equal. If these assumptions/conditions are violated, the resulting P-values and confidence intervals may not be trustworthy³. However, normality is not required for the Wilcoxon signed rank or rank sum tests to produce valid inferences about whether the median of a symmetric population is 0 or whether two samples are drawn from the same population. Thus signs and ranks are only used when the underlying assumptions are violated. I usually perform a visual inspection of the data via histograms, frequency tables. In some instances, if the sample is large enough, and in statistical terms this means greater than 30, then the data is strong enough to withstand non normality.....some 'purists' prefer to 'transform' the data via some geometric transformation to 'normalize' i.e. log 10 transformation etc. That decision you make as the person more intimate with the data.