Using multiple regressions in social sciences research : Some important aspects to be considered

Lay, Yoon Fah (2009) Using multiple regressions in social sciences research : Some important aspects to be considered. Jurnal Kinabalu, Universiti Malaysia Sabah, 13. pp. 73-87.

[img]
Preview
Text
Using multiple regressions in social sciences research.pdf

Download (94kB) | Preview

Abstract

This article discusses one of the most commonly used statistical methods in studying the relationship between dependent and independent variables in social sciences research. The purpose of using multiple regression and types of data suitable for multiple regression analysis are discussed Some important a~pects to be considered when multiple regression analysis is used will be discussed in detail. These aspects are variables selection method (i.e. forward selection. backward elimination and stepwise). Multicollinearity, tolerance, variance inflation factor, influence statistics (DFFIT and DFBETA). Leverage. Cook s distance. standardized regression coefficient (β). coefficient of determination (R²), assumptions such as normality, linearity. homoscedasticity and independence. Figures and tables are illustrated to give better picture of the concepts described.

Item Type: Article
Keyword: Research , statistical methods , variables
Subjects: H Social Sciences > HA Statistics
Department: FACULTY > Faculty of Psychology and Education
Depositing User: NORAINI LABUK -
Date Deposited: 25 Jun 2018 10:16
Last Modified: 25 Jun 2018 10:16
URI: https://eprints.ums.edu.my/id/eprint/20345

Actions (login required)

View Item View Item