In: Statistics and Probability
Prehistoric pottery vessels are usually found as sherds (broken pieces) and are carefully reconstructed if enough sherds can be found. An archaeological study provides data relating x = body diameter in centimeters and y = height in centimeters of prehistoric vessels reconstructed from sherds found at a prehistoric site. The following Minitab printout provides an analysis of the data.
Predictor | Coef | SE Coef | T | P |
Constant | -0.224 | 2.429 | -0.09 | 0.929 |
Diameter | 0.7608 | 0.1471 | 5.29 | 0.003 |
S = 4.07980 R-Sq = 87.5% |
(a) Minitab calls the explanatory variable the predictor variable. Which is the predictor variable, the diameter of the pot or the height?
diameterheight
(b) For the least-squares line ŷ = a +
bx, what is the value of the constant a? What is
the value of the slope b? (Note: The slope is the
coefficient of the predictor variable.) Write the equation of the
least-squares line.
a = |
b = |
ŷ = + x |
(c) The P-value for a two-tailed test corresponding to
each coefficient is listed under P. The t value
corresponding to the coefficient is listed under T. What
is the P-value of the slope?
What are the hypotheses for a two-tailed test of β =
0?
H0: β = 0; H1: β < 0H0: β = 0; H1: β > 0 H0: β ≠ 0; H1: β = 0H0: β < 0; H1: β = 0H0: β = 0; H1: β ≠ 0
Based on the P-value in the printout, do we reject or fail
to reject the null hypothesis for α = 0.01?
Reject the null hypothesis. There is sufficient evidence that β differs from 0.Fail to reject the null hypothesis. There is insufficient evidence that β differs from 0. Fail to reject the null hypothesis. There is sufficient evidence that β differs from 0.Reject the null hypothesis. There is insufficient evidence that β differs from 0.
(d) Recall that the t value and resulting P-value
of the slope b equal the t value and resulting
P-value of the corresponding correlation coefficient
r. To find the value of the sample correlation coefficient
r, take the square root of the "R-Sq" value shown in the
display. What is the value of r? (Round your answer to
three decimal places.)
Consider a two-tailed test for r. Based on the
P-value shown in the Minitab display, is the correlation
coefficient significant at the 1% level of significance?
No, the correlation coefficient is not significant at the α = 0.01 level because the P-value ≤ α.Yes, the correlation coefficient is significant at the α = 0.01 level because the P-value > α. Yes, the correlation coefficient is significant at the α = 0.01 level because the P-value ≤ α.No, the correlation coefficient is not significant at the α = 0.01 level because the P-value > α.
(A) Using the data table, it is clear that the variable under the predictor column is diameter
So, the explanatory or independent variable is diameter
(B) We have the equation of line y = a +bx
where a is intercept and b is the slope
From the data table, value of intercept a = -0.224 and value of slope b = 0.7608
So, we get
a = -0.224
b = 0.7608
this gives us y = -0.224 + 0.7608x
(C) P value for the slope under the P column is given as p = 0.003
So, the p value corresponding to slope or b is 0.003
For two tailed hypothesis,we assume alternate hypothesis to be either less than or greater than the 0 value
So, correct set of hypotheses is and
Based on the p value 0.003, we will reject the null hypothesis at 0.01 level of significance because the p value is less than 0.01, which makes is significant.
We can conclude that there is sufficient evidence to support the statement that beta differs from 0
option A is correct for conclusion
(d) we have r squared = 87.5% or 87.5/100 = 0.875
taking square root, we get
r = (rounded to 3 decimals)
P value shown in the Minitab display is 0.003 which is significant at 1% level of significance because it is less than 0.01. Thus, we can say that the correlation coefficient is significant at 0.01 level of significance because the p value is significant.
option C is correct "yes, the correlation coefficient is significant at alpha = 0.01 level because the p value is "