 So moving forward from parametric testing to another type of testing, that's non-parametric testing. There are many situations and conditions where researchers cannot use parametric tests. You remember there were a lot of assumptions before we use or select any parametric test. For example, distribution, underlying distribution should be normally distributed, then homogeneity of the variance in the groups, etc. So being student of psychology or social sciences, you are confronted with many situations when you are restricted or you cannot use parametric testing. So the alternative choice for us is to non-parametric testing. The main reason to apply the non-parametric test include that the underlying data do not meet the assumptions about the population sample. The population sample size is too small and the analysis data is ordinal or nominal. So if we have our level of measurement ordinal or nominal, or our sample size is very small, or we are unable to meet assumptions of any kind of parametric testing, then our choice would be to test the hypothesis non-parametric testing. Non-parametric statistics does not assume the data is drawn from a normal distribution. Therefore, it is called a distribution-free test. Because we do not first assume that the population that we took data from will be normally distributed. Because the sample size is also small and the sample of some kind that you physically or mentally take your illnesses or data like this where you assume that the underlying data will not be normally distributed. We call this non-parametric assumption-free test or distribution-free test. Non-parametric statistics include that we do everything. Like we do parametric testing, we calculate descriptive statistics, we calculate statistical models, we calculate inference, we draw and we calculate test value as well. Just like we do parametric tests in TEF or KAIS or regression. But the difference is that our model structure is not a priori defined, rather our data tells us how our underlying distribution will be. Non-parametric tests usually involve qualitative data. Remember that the biggest reason when we switch from parametric to non-parametric test is that our data is qualitative or nominal or ordinal categorical data. Our continuous running score does not have data. So most of the time even quantitative data is converted to nominal or ordinal data. There are many situations when we do not have parametric assumptions or our sample size is small. So what we do is that even if our data is quantitative, even if it is on an interval ratio scale, even then we do non-parametric tests and we convert that quantitative data into nominal or ordinal scale data. First, like we do self esteem for students. Self esteem is a running score or quantitative data. If you want to use non-parametric instead of parametric, then you can categorically convert that self esteem. So you can convert that quantitative data into nominal or ordinal data. First, like I make its three categories, low self esteem, medium or high self esteem. So these are some of the introduction and prerequisites when and why we will be using non-parametric statistics.