 Students, we are going to perform a years of distribution analysis through SPSS. Let us see how can we perform this analysis in SPSS. Dear students, in this data set, I have some variables related to the social capital. Online bridging social capital, offline bridging social capital and online bonding social capital and offline bonding social capital. These are the composite variables. We have computed the composite scores from scale and these are the composite scores of these variables. So I had told you that the skewness and catharsis we compute, that is the distribution analysis, can be done only on continuous variables, not on nominal and ordinal variables. So these are the continuous shapes, so let us see what results we have. We will go to the analysis, we will click on descriptive statistics and frequency. We will reset here and we will select our own variables. So these are the four variables that I have selected, online bridging, offline bridging, online bonding and offline bonding social capital. From the statistics, I have taken the mean, median and mod of these three. From dispersion, standard deviation or distribution, skewness and catharsis. I am not going to see the frequency of these variables, I am going to uncheck it by default, we will click OK, so we have this table here. Now if we look here, we have the mean, that is 20.91 of online bridging social capital, median is 21 and the mod is 20. Standard deviation is 4.40 and skewness is minus 0.337 and catharsis is 0.295. Skewness is minus, so this is the negatively skewed data, because our median value is bigger than our mean value and this is why it is negatively skewed. Skewness's level is not so high that it is not between 0 to 1. The catharsis that we talk about is positive, so we will say it is going towards leptocortics and mesocortics, but because it is between 0 to 1, it is tolerable. In offline bridging social capital, we have mean 22, median 23 and mod 24. Now the value of the mod is bigger than the mean and median, this is why the data is negatively skewed and its value is 0.822, so we can say that it is tolerable because it is smaller than 1. The catharsis value is positive, 0.983, it is also tolerable, but it is more towards leptocortics, i.e. the data is taller. The mean of online bridging social capital is 16, median 17 and mod is 18, so this is why it is negatively skewed, because the mean and median value of the mod is bigger. The catharsis value here is minus 0.156, so we can say that the data is flattening, i.e. it is becoming a leptocortics symmetry of its shape. In offline bridging social capital, the value of the mod is more than the mean and median, this is why it is also negatively skewed, but it is less than 1, that is why we can say that it is tolerable. The catharsis value is also minus 0.159, so the shape of it is becoming a leptocortics. So we have a leptocortics shape in 2, and it is flattening in 2, our shape is going towards the leptocortics. So we can say that in these 4 variables that we have seen in this dataset, the catharsis and skewness values have come from this, the data is a little skewed, both on the positive side and the negative side for these 4 variables, and the catharsis value is also there, but it is tolerable, because none of these values have come from 1, so we can say that this data is okay for the parametric testing, because if your values are okay for normal distribution, then you apply parametric tests. If their values are the highest of distribution, they are telling that your normal distribution has been violated, i.e. the data is very skewed, and there is a lot of tallness and flatness, then we do not apply parametric tests. So it is important to check the shape of the distribution, so that further analysis is required. Along with this, we have done all 3 things in our descriptive analysis, features of data, measures of distribution, measures of dispersion and measures of central tendency. Along with this, our section of analysis is completed. Now we will see in the next module, that in our descriptive analysis, can we bivariate it or multivariate it? And how can we do that? So we will discuss this in the next module.