Multiple Regressions: Statistical Methods Using SPSS

Introduction

Multiple regression analysis provides scope for researchers to conduct in-depth analyses about two variables, one is dependent and another is independent. Through Multiple regression analysis, it is hereby possible for researchers and statisticians to identify the interlink between the outcome (the dependent variable) and several predictor variables which are considered as independent variables in the data set. The article is helpful to understand the multiple regression analysis and explore the easy way to conduct the analysis after data handling and sorting through MS Excel and SPSS. The relationship between the independent variable and dependent variable can be explored and evaluated through multiple regression analysis. The researchers can utilise SPSS for performing the multiple regression analysis in order to progress in the study by identifying the impact of the independent variables on its dependent variables, in order to draw the final conclusion.

Multiple regression analysis

The formula for the multiple regression analysis is,

Y = B1X1 + B2X2 + … + BnXn + C

Where,

Y is the dependent variable

X is the independent variable

b is an unknown parameter

C is the constant term

The multiple regression method provides scope to researchers to include two or more independent variables in the data set to identify the impacts of all the independent variables on its dependent variable. The inclusion of several independent variables in the data set further helps to analyse the influential effects of the outcomes. for example, doctors try to analyse high blood pressure patients and diagnose their height, weight, age, hours of exercise per week, and lifestyle. Hence, in this particular data set, blood pressure is the dependent variable, and other body weights, the height of the patient, lifestyle, eating habits, and hours of exercise per week are considered as independent variables, where the impacts of weight, heights, and hours of exercise on the blood pressure of the individual can be diagnosed. Here, the sign of the coefficient determines the inert link between the variables. The sign on the coefficient (positive or negative) gives the direction of the effect, such that if the coefficient is positive, there is a positive correlation between the dependent and independent variables. If the increase in the independent variable will decrease the dependent one, the correlation coefficient is negative. Multiple regression analysis is utilised widely by researchers and statisticians in order to analyse the interrelationship between two or more independent variables and one dependent variable. The main uses of such multiple regression analysis are forecasting future activities, time series modeling, and finding the cause and effect relationship between variables. Hence, the impacts of one variable to another variable can be analysed through multiple regression analysis, where it provides a scope to consider two or more independent variables in the data set that affect the dependent variable. Some of the independent variables improve the modeling where it is possible for the researchers to reduce the error term and identify the appropriate effects of the independent variables on the dependent one.

Statistical methods for multiple regressions

The assumptions to perform multiple regression are such as the dependent variable is measured on a continuous scale that meets the criteria like revision time, intelligence, exam performance, weight, etc., and additionally, two or more independent variables should be there that can be either continuous or categorical. The examples of nominal variables are gender, ethnicity. Physical activity level professional etc., are considered as independent variables in the data set. The researchers should have independence of observations and along with this, there needs to be a linear relationship between (a) the dependent variable and each of your independent variables, and also (b) the dependent variable and the independent variables collectively. The data needs to show homoscedasticity and must not show multi-collinear. Checking the existing error whether it is normally distributed or not is also examined efficiently through the regression analysis. After data handling and putting the collected data in SPSS, the researcher clicks on the options, analyse, regression, and linear. In the dialogue box, the dependent and independent variables must be chosen by the researchers.

By clicking the statistics, there will be a linear regression statistics dialogue box, where the confidence interval must be set at 95%. After continuing the operations, there will be multiple regression analyses, on the basis of which, the researchers can draw final conclusions. The multiple correlation coefficient is determined by the value of R, where researchers can analyse whether the data fit the regression model. It is also important to set a confidence interval at 95% to explore the value of r. R-squared hereby measures the strength of the relationship between your model and the dependent variable on a convenient 0 – 100% scale. For example, an r-squared of 60% reveals that 60% of the data fit the regression model and there is a correlation between the dependent and significant independent variables. The standards for a good R-Squared reading can be much higher, such as 0.9 or above. SPSS is hereby an effective statistical tool to perform multiple regression analysis and explore R values and t statistics for further critical evaluation.

Summary

Multiple regression analysis is hereby conducted through MS Excel and the SPSS tools for exploring the impacts of several independent variables on one dependent variable in the data set. Researchers must consider the two or more independent variables to utilise this multiple regression tool in SPSS software for conducting in-depth research. The master thesis can also be conducted through exploring R values, where it is possible for the researchers to identify interlinks between the variables in the data set for further evaluation. Multivariate linear regression is a widely used machine learning algorithm, where researchers are able to identify interlinks between the variables and explore the positive or negative correlation between them.