Checking whether All Assumptions Were Tested

There are key assumptions that should be met by data or variables before carrying out multiple regression analysis in SPSS. Some of these assumptions include absence of multicollinearity in the data (Mohd, 2016), the presence of homoscedasticity in the data (Srinivasan & Lohith, 2017), a linear relationship between the independent and the dependent variables (Forister & Blessing, 2015), and independence of residuals or observations (Anderson, Sweeney, Williams, Camm, & Cochran, 2017).

Is your time best spent reading someone else’s essay? Get a 100% original essay FROM A CERTIFIED WRITER!

Order now

Other assumptions include normality of data, undue influence, independence of error, and linearity (Takahashi, Inoue, & Trend, 2016). Out of the seven assumptions of multiple regression analysis, my colleague checked the following assumptions: linearity, homoscedasticity, independence of error, lack of multicollinearity, and whether there are normally distributed residuals.

Additional Resources That Are Useful in Addressing Diagnostic Issues

I want to recommend to my colleague some online resources which are useful in understanding how to check the assumptions of multiple regression analysis. One of these resources is Ernst and Albers' (2017) article. The article addresses four OLS-assumptions (Independence of Errors, Homoscedasticity, Normality, and Linearity). More importantly, Ernst and Albers (2017) address the consequences of violations of assumptions as well as common misconceptions about assumptions.

Another online resource that provides a step-by-step guide of testing the assumptions of multiple regression analysis is The Open University Faculty of Arts and Social Sciences' (n.d.) tutorial. The tutorial provides a practical step-by-step guide of checking the following assumptions of multiple regression analysis: linearity, absence of multicollinearity in the data, homoscedasticity, undue influence, independence of residuals, the variance of the residuals, normal distribution of residuals, and influential cases biasing the model (The Open University Faculty of Arts and Social Sciences, n.d.).

References

Anderson, D. R., Sweeney, D. J., Williams, T. A., Camm, J. D., & Cochran, J. J. (2017). Statistics for business & economics, revised. Boston, MA: Cengage Learning.

Ernst, A. F., & Albers, C. J. (2017). Regression assumptions in clinical psychology research practice-a systematic review of common misconceptions. Retrieved from https://peerj.com/articles/3323/

Forister, J. G., & Blessing, J. D. (2015). Introduction to research and medical literature for health professionals. Burlington, MA: Jones & Bartlett Publishers.

Mohd, S., Norazah. (2016). Handbook of research on leveraging consumer psychology for effective customer engagement. Pennsylvania, PA: IGI Global.

Srinivasan, R., & Lohith, C. P. (2017). Strategic marketing and innovation for Indian MSMES. New York, NY: Springer.

Takahashi, S., Inoue, I., & Trend, C. L. (2016). The manga guide to regression analysis. San Francisco, CA: No Starch Press.

The Open University Faculty of Arts and Social Sciences. (n.d.). Assumptions of Multiple Regression. Retrieved from http://www.open.ac.uk/socialsciences/spsstutorial/files/tutorials/assumptions.pdf