At the heart of quantitative research is regression analysis. A regression model simply measures the association of two variables - variation: the dependent variable's observed value from its mean in relation to the independent variables observed value from its predicted value. I refer to this as 'variation of numbers in a column of a spreadsheet'. The regression calculation does not distinguish the meaning of these numbers. It is a researcher who performs the task of giving meaning to the output of a regression analysis. In this discussion, analyze the role of a researcher in performing regression analysis and how he or she gives meaning to the regression output.
A researcher usually approaches a problem with some knowledge of how two groups of data relate to one another. Either this knowledge is from experience or from an observation of the results of an experiment. The researcher also typically has some sense of which group of data may depend on the other. For example, one variable may be exposure to sunlight. Another variable may be vitamin D levels in the body. A reasonable experiment may be to test how exposure to sunlight affects vitamin D levels. In this case, vitamin D levels would be the dependent variable and exposure to sunlight would be the independent variable. One could run an analysis the other way (i.e., how do vitamin D levels affect exposure to sunlight?), but such an analysis would not make much sense in the real world. Therefore, the researcher using good judgment must choose a ...
The solution analyzes the role of a researcher in regression analysis.