Lesson 12: Regression NAME Lab Assignment Answer the following questions showing all work. For questions that require Minitab Express, include the appropriate output (copy + paste) along with an explanation. Use an alpha level of .05 unless otherwise specified. 1. Use the file Class Survey (in the data sets folder for Minitab format) to answer the following questions. Data from college students aged 18-24. (20 points) A. Create a scatterplot with weight on the Y-axis and height on the Xaxis. Copy and Paste your scatterplot. Summary Statistics Variabl e N Mean StDev Minimum Maximum Weight 225 152.560 30.844 98.000 290.000 Height 225 67.6533 5.9059 52.0000 106.000 Pearson correlation of Height and Weight = 0.371823 P-Value = <0.0001 Analysis of Variance Source DF Adj SS Adj MS F-Value P-Value Regression 1 1080.16 1080.16 35.78 <0.0001 Weight 1 1080.16 1080.16 35.78 <0.0001 Error 223 6732.80 30.19 Lack-of-Fit 57 1640.27 28.78 0.94 0.6014 1 Lesson 12: Regression NAME Lab Assignment Pure Error 166 5092.53 30.68 Total 224 7812.96 B. Describe the scatterplot that you made in part A in terms of direction, shape, strength, and outliers. C. Would it be appropriate to compute a Pearson's r as a measure of the relationship between these two variables? Why or why not? Yes, it would be appropriate to compute a Pearson's r as a measure of the relationship between the two variables because there is a linear association. D. Compute the correlation (Pearson's r) for the relationship between weight and height and use the five-step hypothesis testing procedure to determine if the correlation is statistically significant. (5 pts of 20) 1) Check necessary assumptions and write null and alternative hypothesis Ho: p=0 Ha: p0 2) Calculate appropriate test statistic using software Correlation 3) Determine P-value 4) Decide between null and alternative hypothesis 5) State real world conclusion E. How would you explain the results that you found in parts A through D to a friend with no knowledge of statistics? F. Use the five-step hypothesis testing procedure to determine if height (in inches) is a statistically significant predictor of weight (in pounds). Assume that all assumptions have been met. (5 pts of 20) A value is statistically significant if it is less than alpha, or .05 in this case. The height is therefore, statistically significant since its o-value is less than .05. G. Harry is an adult. He is 65 inches tall. Use your regression equation from part F to predict Harry's weight. Regression Equation Height = 56.792 2 Lesson 12: Regression NAME Lab Assignment + 0.07120 Weight H. Harry's actual weight is 154 lbs. What is Harry's residual? I. Harry has no knowledge of statistics. How would you explain what his residual means to him? J. Levi is 7 years old. Would it be appropriate to use your regression model from part F to compute his residual? Why or why not? K. Interpret the slope in part F in the context of this problem. 3