See data file attached.
Management at the Texas Christian University physical plant is interested in reducing the average time to completion or routine work orders. The time to complete is defined as the difference between the date of receipt of a work order and the date closing information is entered. The number of labor hours charged to each week order and the cost of materials are two variables believed to be related to the time to completion of work order. Management wants to know if there is any difference in the time to completion of the work order, on average, for different types of buildings. Buildings are classified into four types of the TCU campus: residence halls, atheltic, academic, and administrative. In answering the question, take into account the possible effect of labor hours charged and material cost. The data for a random sample of 72 work orders (chosen form population of 11,720) are is file attached. The variables are as follows:
y= DAYS = Number of days to complete each work order
x1=HOURS = Number of hours of labor charged to each work order
x2=MATERIAL = cost of material charged to each work order
x3 = BUILDING = 1 for residence hall
2 for athletic building
3 for academic building
4 for administrative buildings
Please see the attached files
In order to examine if there is any significant difference in the completion time of the work order, on average, for different types of buildings (in addition to the variables, number of labor hours charged and cost of material charge), we have to fit a regression model by taking number of days to complete as the dependent variable and the other variables as independent variables.
But, here the problem is, building type is a categorical variable having four different categories: residence hall, athletic building, academic building and administrative buildings. A categorical variable cannot be directly included in a regression model. It is possible, however, to include dichotomous (binary) variables in a regression model, since the ordinary least square procedure treats this as a continuous variable that can take values between zero and one. Hence, the technique proposed is to introduce sufficient number of dummy indicator variables to ...
The solution uses the indicator variables to reduce average task times.