STATISTICAL TUTORIAL CHANNEL RESEARCH
Find out which is the best statistical test for your research!
IN ADDITION TO VERIFYING A POSSIBLE RELATIONSHIP OF YOUR VARIABLES, YOU ALSO WANT TO VERIFY IF YOUR INDEPENDENT VARIABLE (PREDITOR) CAN SIGNIFICANTLY INFLUENCE A DEPENDENT VARIABLE (OUTCOME), IE WANT TO DO A REGRESSION ANALYSIS ?_cc781905-5cde-3194-bb3b-94-bb3b-94 136bad5cf58d_
?
In this case, the researcher just wants to know if two or more variables are related to each other. But this analysis must always be done with the variables paired two by two.
This analysis is indicated to verify the association/correlation of variables. The term correlation means relationship in both senses: it describes the association between two variables, making no judgment about whether one is the cause or consequence of the other. The purpose of this association is to study how much two variables change together, without, however, implying a cause and effect relationship of one variable over the other.
Correlation is used when you want to study how consistently two variables change together. When this occurs, statisticians say there is a correlation or covariation, the direction and magnitude of which can be quantified.
The classic example is the researcher who wants to study whether the variables weight and height of a population of 12 adult men are related. In this case, the researcher just wants to know if the taller the person is, the higher or lower their weight values should be. The researcher does not expect a cause and effect relationship, nor does he expect to predict values. This test tells if the variables tested are significantly related and if so, it also shows the strength of this relationship.