 So, now let's move on to SPSS and solve all the things that we have talked about what are the assumptions of the simple linear regression where we have one independent variable and then the multiple regression where we have more than one independent variable. To starting with a simple linear regression, the variable we want to predict is called the dependent variable and the predictor is of course the independent variable. For simple linear regression, dependent variable should be measured in the continuous level we have talked about before that our dependent variable should always be on the interval ratio scale, continuous, quantitative running score. Independent variable could be continuous or quantitative or could be categorical having two categories or more. If the dependent variable is categorical, if we have a dependent variable categorical then we can't use linear regression or radical regression, in fact, what we will do is, other technique which is non-parametric, we apply logistic regression on it. So, remember logistic regression is also a very good tool in statistics when the dependent variable is categorical, then we solve it with logistic regression. There are many examples in our business and social studies where our dependent variable is continuous, for example, success of failure or good employee or bad employee, pass or fail. For example, there are so many criteria, there are tons of independent variable and you want to see which is predicting success. So, then you move on to logistic regression. So, this is an example and we will feed the data in the SPSS, we will check the assumptions as well and then we will look at the analysis. In this, we have added all kinds of virtual things in the SPSS, but this is an example where there are two variables, one is the dependent variable which is income and then the independent variable which is weight. To simplify the weight variable, women are classified into five categories. So, weight is categorical variable which is divided into one to five. So, weight can be divided into five categories, one is the thinnest and five is the heaviest. So, find the linear regression. What is the regression account for significant portion of the variance? So, we have to find out what is your weight, whether any variance is explaining your income or not? So, we have to check on 0.05. Let's move on to SPSS. I have already added this data in the WES. The same data I have shown you, so these are the 10 participants. Participants, first we have the weight variable, which we categorically have done in 5 categories and then we have the income. So, income is a dependent variable, it is continuous running score and this is our independent variable. So, we will go to analyze and then we will go to general linear regression and then we will go to linear regression. So, you remember that we have to go to linear regression, all these other functions are also very important in this but still we will stick only to the linear regression. So, if you have gone to linear regression, you have this type of window pop-up. We have two variables because it is a simple linear regression, our dependent variable is our income and our independent variable is our weight. We will set this. These are all the types that I have just talked about. There is an enter method, your stepwise, remove, backward and forward. So, enter means that it is simple, simultaneous or hierarchical when we press the next button. So, in simple regression, we don't need to go next because we have one variable. We already have a model fit in the statistics. We don't need to change R square because if we have hierarchical regression where we first put a variable in a block, then it is very important to tell us how much change is taking place in the next block. Descriptives are very important because you get a lot of idea of overall data. What are the descriptives, means, standard deviation, standard error, minimum, maximum data. It is not necessary to do confidence intervals. If you have to do bootstrapping, then it is necessary. Let there be part and partial correlation. Colineality diagnosis too because we don't have more than one variable. So, we don't need it. We also said that we check our auto-correlations. We can do it because it tells us the consistency of the residual variance in your dependent variable. So, we will continue after that. These are our plots. It is very important because these are our scatter plots. This is the plot of error or residual term. What we talked about in your data, the homoscedasticity of the heteroscedasticity, we can basically check with it. So, z residual will come on our y and z predicted will come on our x variable. So, your predicted or residual will come on it. So, predicted score and then error term are with us. You can say that we should do normal probability plot. Normal probability plot will tell us that our normality assumption, which we talked about in regression assumptions, is meeting it or not. So, you will continue. After that, the save means that what we showed you by hand, how we take out predicted values. So, we take out predicted values for each axis. So, if you need a column for your predicted values, for example, if the weight is in category 4, then how much will be the income? Then you can press the save button in it. But you don't need it now. In the options, by default, your probability or what? Confidence interval check. You don't need that. You don't need bootstrapping also. So, this is a simple regression where we need just some statistics plot and that's it. And we will press the OK button. And we will have this output. So, now, in the next scope, in the power point, I will explain one by one each table.