This is well known fact that the success of social science research heavily depends upon the selection of research tools and its effective utilization. Researchers often come across the situations where they want to study the impact of one variable on the other variable viz. impact of income on expenditure. Although we have freedom to select research tools for multivariate analysis as wide range of research tools are available, multiple regression analysis allows us to determine the effect of more than one independent variable on dependent variable. This term paper talks about the concept of multiple regression analysis, its assumptions, application, and its limitations to the social science research. The paper also briefs about various statistics associated with multiple regression analysis.
Table of Contents
The document does not contain a table of contents.
Objectives and Topics
This term paper aims to explain multiple regression analysis as a tool for social science research. It addresses how relationships between a dependent variable and several independent variables can be expressed and assessed, while introducing the method’s assumptions, applications, associated statistics and limitations.
- Simple and multiple regression models
- Prediction and relationships between variables
- Scatter diagrams and least squares
- Assumptions underlying multiple regression
- Regression coefficients and statistical measures
- Limitations of regression analysis
Excerpt from the Book
Overview
Regression analysis is a statistical technique to investigate the relationships between quantitative variables. In some situation, researchers are interested to determine the underlying effect of one variable on another variable viz. effect of income on expenditure or effect of changes in money supply on the rate of inflation. At the same time, the researchers also assess the “statistical significance” of the estimated relationships, that is, the degree of confidence that the true relationship is close to the estimated relationship.
Regression analysis is a powerful statistical technique that identifies the association between two or more quantitative variables: a dependent variable, whose value is to be predicted, and an independent or explanatory variable (or variables), about which significant amount of knowledge is available. This statistical tool is used to develop the equation that represents the relationship between the variables. A simple regression analysis can show that the relation between an independent variable and a dependent variable is linear, using the simple linear regression equation. Multiple regression analysis provides an equation that predicts dependent variable from two or more independent variables. In other words, it can be said that multiple regression involves a single dependent variable and two or more independent variables, while simple regression model involves one dependent variable and one independent variable.
Regression analysis is concerned with the nature as well as the degree of association between variables. Although the independent variables may explain the variation in the dependent variable, it does not necessarily imply causation. At the time of using multiple regression analysis in psychology, many researchers use the term “independent variables” to identify those variables that they think will influence some other “dependent variable”. Research literature also uses the term “predictor variables” for those variables that may be useful in predicting the scores on another variable that we call the “criterion variable”.
Section Summaries
Overview: The paper introduces regression as a means of studying relationships between quantitative variables. It distinguishes simple from multiple regression and cautions that association does not necessarily imply causation.
Multiple Regression Equation: The general equation is presented with its dependent variable, independent variables, constants and error term.
Using Multiple Regression Analysis: This section describes uses including prediction, assessing explained variation, controlling for variables and testing hypotheses.
Plotting the Scatter Diagram: The paper discusses scatter diagrams as visual representations of relationships and introduces least-squares line fitting and residuals.
Assumptions for Multiple Regression Analysis: The section lists assumptions concerning linearity, residuals, error terms and conditional distributions.
Performing Multiple Regression Analysis: The general model is revisited, and raw and standardized regression coefficients are explained.
Some Statistics associated with Multiple Regression Analysis: The paper introduces measures and tests including R², adjusted R², regression coefficients, the F-test, the t-statistic and sums of squared errors.
Limitations of Multiple Regression Analysis: The discussion considers causal interpretation, circular relationships, data requirements and violations of model assumptions.
Keywords
multiple regression, social science research, dependent variable, independent variables, prediction, linear regression, scatter diagram, least squares, residuals, regression assumptions, regression coefficients, coefficient of determination, hypothesis testing
Frequently Asked Questions
What is this paper about?
It introduces multiple regression analysis and its use in social science research.
What are its central topics?
The paper covers regression equations, applications, scatter diagrams, least squares, assumptions, statistical measures and limitations.
What is the paper’s primary objective?
Its objective is to explain the concept and use of multiple regression, including its assumptions and limitations. It does not state a formal research question.
What scientific method does the paper use?
It is an explanatory discussion of multiple regression analysis; it does not report an original empirical study.
What does the main body cover?
It presents the regression equation and its uses, then discusses scatter diagrams, least-squares fitting, model assumptions, coefficients and associated statistics.
Which keywords characterize the work?
Key terms include multiple regression, independent variables, dependent variable, least squares, residuals, regression coefficients and coefficient of determination.
How does the paper distinguish simple from multiple regression?
It describes simple regression as involving one independent variable and multiple regression as involving two or more independent variables.
What does the paper say a scatter diagram can show?
It can visually display the pattern of association between two variables, but it cannot by itself establish that one causes the other.
Which statistics are introduced?
They include R², adjusted R², regression coefficients, the F-test, standard error, the t-statistic and the sum of squared errors.
- Quote paper
- Kunal Gaurav (Author), 2010, Multiple Regression Analysis: Key To Social Science Research, Munich, GRIN Verlag, https://www.grin.com/document/183645