Math 117 Computer Lab Assignment – Regression Analysis

(Due at the beginning of lab next week – 2/1 or 2/3)

Consider the data in the BodyDimensons.sav file (in lab, you should have copied this file to your account). In a 2003 study, weight (in kilograms), age (in years), and multiple body circumference measurements (in centimeters) were measured on 260 physically active women.

Using PASW to aid your analysis (don’t do these by hand), answer the following questions. It might be easiest for you to cut-and-paste PASW output into Word and type your comments around that output—that is, you might want to create a Word document for this assignment, rather than writing directly on this sheet.

  1. You want to predict a woman’s weight (in kg). Using the  value as a guide, determine the best individual predictor variable of weight. (Consider only models with a single explanatory/predictor variable.)

 

    1. What is the best predictor variable? (Include a well-labeled scatterplot to show the relationship between this predictor variable and weight.)

 

______________________________________________________________________________

 

    1. For the predictor you chose, what is the equation of the regression line?

 

______________________________________________________________________________

 

    1. Based on your regression line given in part b, carefully interpret—in words specific to these variables—the numerical value (not just the sign) of the slope of the regression line.  

 

______________________________________________________________________________

 

______________________________________________________________________________

 

______________________________________________________________________________

 

    1. What is the value of  for your model? Furthermore, what is the interpretation (in words specific to this situation) of this  value?

 

______________________________________________________________________________

 

______________________________________________________________________________

 

______________________________________________________________________________

 

    1. Create a residual plot for your regression model (and turn in this graph). Based on the residual plot, is there any indication that a straight line is not the best way to summarize the relationship in the data? Why or why not?

 

______________________________________________________________________________

 

______________________________________________________________________________

 

______________________________________________________________________________

 

______________________________________________________________________________