 Aur so is now we are going to talk about multiple linear regression multiple linear regression is actually an extension of simple linear regression but there is one basic differenceination mere ka pass a quick independent variable with another independent variable more than one independent variables it can be 2 and it can be 10 or it can be more than 10 in multiple linear regression we have to check various assumptions just like that the simple linear regression the first assumption for the multiple linear regression is that your dependent variable should be made at continuous level it should be an indexed score it should be a composite score when we talk about the independent variables majority of your independent variable should be in continuous in nature and there may be some variables that can be included in the model and they may be nominal variables like gender and they may be variables like sex and there may be variables like religion or any possibility of yes no before performing multiple linear regression we have to check the possibility of higher outliers in our data set we have talked in detail about how outliers affect your linearity and relationship there must be a linear relationship between IVs and dependent variables we have also talked in the simple linear regression that the simple linear regression it checks a linear relationship it doesn't check your parabolic relationship it doesn't check the exponential relationship so for multiple linear regression it is very important to fulfill these assumptions that your IVs and dependent variables should be in linear in nature your data should have multivariate normality i.e. data of predicted residuals or standardized residuals should have a line fit multivariate normality shouldn't be between independent variables now we can discuss multivariate normality in detail and there should be homosidacity in the data set data should be scattered in the center and there should be no relationship between the values of x and the error terms if we talk about this procedure in SPSS software we go to analyze we select linear from regression and with our dependent variables we add your set of independent variables in this module our dependent variable is online bridging social capital and our independence are various forms of motivations to use social media in this online self-declosure making new social ties, maintaining existing social ties seeking and sharing information, self-expression we have added all the motivations in this as independent variable to predict the variance of online bridging social capital when you click on its options by default you have estimate and model fit the box of fit will be checked along with this you have to check your residuals along with this auto correlation you have to check your Durban Watson you have to talk about model fit and this is it we can see the change in our scale along with the descriptives along with the solidarity diagnostic the assumption of linear regression is that when we conduct multiple linear regression the independent variables should not be auto correlation the correlation value should not be very high so we will see this in more detail from the plots your by default the open dialog box in this you have to do that you have to bring z predicted on x and z residuals along with this you have to check the normality probability plot after this we will continue along with these options when we do OK we will have the analysis run and it will come to the out box which we call output view in the next module we will talk to them and will relate the results with assumptions which assumptions are fulfilled and what are our results