Question

In what follows use any of the following tests/procedures: Regression, confidence intervals, one-sided t-test, or two-sided t-test. All the procedures should be done with 5% P-value or 95% confidence interval.

Open Weight_vs_IQ data. SETUP: Common sense dictates that a person’s IQ and Weight should not be related. However, one never knows until one examines the data. Given the data (IQ and weight for 29 students) your job is to check if the common sense assumption is reasonable or maybe it is not.

13. What test/procedure did you perform?

a. Regression

b. Two-sided t-test

c. One-sided t-test

d. ??Confidence interval

14. What is the P-value/margin of error?

a. 0.349971

b. 8.2E-06

c. 0.211936

d. 0.155968

e. ??None of these

15. Statistical interpretation

a. Since P-value is large we can confirm the common sense assumption that IQ and Weight are not related.

b. Since P-value is large we cannot claim that averages are different .

c. Since P-value is large we cannot claim that IQ and Weight are related..

d. None of these

16. Conclusion

a. I have no statistical evidence to claim that IQ and Weight are related.

b. I am confident that IQ and Weight are related.

c. I am confident that IQ and Weight are not related.

Weight  IQ
88      99
112     103
107     97
112     99
80      100
86      90
103     118
121     79
116     80
82      88
65      82
107     103
100     116
91      112
110     92
113     115
105     100
108     94
90      118
73      106
102     126
99      95
80      119
114     116
90      79
95      92
106     78
86      111
106     80

Answer 1

13. The aim is to check if the commonsense assumption that person's IQ and Weight should not be related is reasonable or not, that is the requirement is to check the relationship between the two variables. Hence, the correct procedure is a. Regression. 14. The Regression output of the data is as follows: Regression Analysis: Weight versus IQ Analysis of Variance Source DF Adj ss Adj MS F-Value P-value Regression 1 21.2121.210.10 0.750 1 21.21 21.21 0.10 0.750 Error 27 5524.93 204.63 Lack-of-Fit 19 4086.43 215.08 1.20 0.417 Pure Error 8 1438.50 179.81 Total 28 5546.14 Model Summary S R -sq R-sq(adj) R-sq(pred) 14.3048 0.38% 0.00% 0.00% Coefficients Term Coef SE Coef Constant 104.3 19.1 1Q -0.061 0.190 T-Value P-value VIF 5.46 0.000 -0.32 0.750 1.00 Regression Equation Weight = 104.3 -0.061 IQ Fits and Diagnostics for Unusual Observations Obs Weight Fit Resid Std Resid 11 65.00 99.25 -34.25 -2.51 R R Large residual

In the Coefficients table, the P-value for IQ is 0.750, therefore, the correct answer is None of these. 15. The correct answer is c. Since P-value is large we cannot claim that the IQ and Weight are related. 16. The conclusion is a. I have no statistical evidence to claim that IQ and Weight are related.