As you can see below, when I include both number of bedrooms and sq feet in the same model, number of bedrooms all of a sudden become not-related to rent.
Square feet is related to rent when number of bedrooms is present in the model (the p value is .05 which is less than or equal to alpha i.e. 05).
Number of bedrooms is nt related to rent when square feet is present in the model (the p value is higher than alpha i.e. .05).
The reason being that both sq feet and number of bedrooms are highly correlated. The correlation between sq feet and number of bedrooms is .68 (almost close to .70 – which is the thumb rule of finding out if the correlation is high or not). When the correlation is so high, what happens is that the two variables invariably are sharing same story. In this case since square feet and rent have a higher correlation than number of bedrooms and rent, thats why square feet becomes significantly related to rent, and number of bedrooms is not related to rent in this case.
Refer to the correlation table below:
