Multiple regression analysis is a very powerful statistical tool that provides a mathematical model of some physiological situation in life. The model can include simultaneously both quantitative and qualitative factors in the prediction of a continuous response variable. Quantitative factors are generally modeled by using either polynomial, exponential, or logarithmic equations, whereas qualitative factors are modeled by using dummy variables. The researcher must determine the factors to include in the model, their possible interactions, and the appropriate form of the model, and then the least squares theory is used to find the best equation of that form. Other statistical analyses, such as discriminant analysis and log linear models, are available for situations in which the response variable is non-numerical.
Get full access to this article
View all access options for this article.
References
1.
MarksRG. Principles of statistics in respiratory research: Part 5. An introduction to statistical analysis. Respir Care1980;25:649–53.
2.
MarksRG. Principles of statistics in respiratory research: Part 3. Research design: The concept of repeated measures and estimating the sample size. Respir Care1980;25:448–51.
3.
KleinbaumDG, KupperLL. Applied regression analysis and other multivariable methods. North Scituate MA: Duxbury Press, 1978.
4.
McClaveJT, DietrichFG. Statistics. San Francisco: Dellen Publishing, 1979.
5.
AgrestiA, AgrestiBF. Statistical methods for the social sciences. San Francisco: Dellen Publishing, 1979.
6.
OttL.An introduction to statistical methods and data analysis. North Scituate MA: Duxbury Press, 1977.
7.
MarksRG. Principles of statistics in respiratory research: Part 6. Hypothesis testing. Respir Care1980;25:745–50.
