Resolved: Causal or correlated relationships?
At 1:02: "A model used for quantifying causal relationships among the different variables included in your analysis". Should it not be "correlated relationships"?
Hi Clifford!
Thanks for reaching out.
We normally quantify a relationship, or a correlation between two variables. Ideally, we'd like to find out the movement of which of the variables causes the movement of the other. This is not always possible to be done and we can use various tools to find the cause, but regression analysis is namely a tool we can use to find out the cause for a certain phenomenon.
Hope this helps.
Best,
Martin
Hello Clifford,
Despite both correlation and regression analyses provide a view about the relationship between two variables or more, they are distinct from each other.
Correlation analysis aims to identify if there is a relationship between two variables or not. If there is, it assesses the strength and direction of this relationship, using the correlation coefficient ranging from -1 to +1, with zero indicates no correlation; -1 indicates negative correlation; +1 indicates positive correlation; and the distance from zero to -1 or +1 refers to the strength of this correlation. So, as you can see the correlation analysis can only tell us if there is a relationship or not, but cannot tell us if one variable is dependent on the another in its value.
On the other hand, regression analysis is primarily concerned about the causal relationships. As you have read, it aims to identify if that relationship between two variables, elicited by the correlation analysis, is causal in nature or not (i.e., one variable is dependent on the another or not).
Kind regrards,
Kyrillos