where, X1 is worker density defined as number of workers on main shift per 1000 m2 of GFA and X2 is a dummy variable, which is 1 if the hotel is a 3-star development and 0 if it is a higher class hotel (4- or 5-star).
In performing regression analysis, we have made certain assumptions about the errors; the usual assumptions are that the errors are independent, have zero mean, a constant variance and follow a normal distribution (Draper and Smith, 1981). A close examination of the residual plots revealed no characteristics which will lead to the violation of the above listed assumptions.