Question

1.Why we recommend to calculate the correlation coefficient factor before applying the Regression Analysis? if you found the r as 0.25, what is your next interpret on next action?

2.You know that adjusted r2 is one the primary outputs for each Regression model. Suppose that we developed two different models (x is Skill and Y is team performance) with following r2 :

Model 2: What does the adjusted r2 = 0.78 mean? Explain.

Model 3: What does the adjusted r2 = 0.97 mean? Explain.

Which of these two models do you think is the best predictor of team performance? Explain.

Answer 1

1. Calculation of correlation coefficient before regression analysis is necessary so as to understand how useful the regression analysis is going to be. Suppose if the correlation coefficient is 0.25, which is a very small value close to 0, there's not much use of linear regression because coefficient close to 0 will indicate that most of the actual data points do not lie on a straight line which is why the estimated regression line will also not estimate the data points correctly.

2.Adjusted R square = 0.97 means 97% of the total variation of Y is explained by the fitted regression model on Y, keeping in consideration the number of variables and observations used for the regression model.

More the Adjusted R square, better is the model since better is the total variation in Y explained.

So, model with adjusted R square = 0.97 is better than the model with adjusted R square = 0.78.