CEG2002:
Statistics for Civil Engineers
Practical 2
Instructions
1. Answer all questions.
2. This practical is not assessed but you should work through all
questions, saving your work at the end of the session for future use.
You should write your solutions to the
following questions in a Microsoft WORD document (this is how the computing
part of the Statistics assignment will have to be written up). Remember to include a title at the top of
the page e.g. CEG2002 Practical 2.
Dont copy and paste the entire Minitab output into your solutions
just the relevant bits! And dont forget
to save your work!
http://www.staff.ncl.ac.uk/j.q.shi/teaching/CEG2002/
Click the name Barcelona_gothic.MTW
to download the data file (or open the data file directly). In Minitab,
click on File and Open Worksheet, and then open the data
file you just saved.
In Column C1 of the
Minitab worksheet you should now see the data, which show the lead content (in mg
per litre) of 10 jars of water taken from the public supply in the Gothic
Quarter of Barcelona (it will also be discussed in Tutorial 2).
(a) Test the null hypothesis that the average lead content in this
part of the city is 50 mg per litre, by clicking on Stat
Basic Statistics 1-Sample t. Enter C1
in Samples in columns, click Perform hypothesis test and enter 50 as
the Hypothesized mean. Click OK. Look at the output in the Session window, and
then answer the following questions.
i. What is the test statistic as calculated by Minitab?
ii. What is the exact p-value for this hypothesis test? Interpret this p-value and form your conclusions.
iii. Write down the 95% confidence interval for the population mean
lead content of jars of water in collected in the Gothic Quarter of
Barcelona.
(b) Recall the assumptions of independence and Normality which
underlie the test you performed in part (a).
i. Produce a histogram and a Box-and-Whisker plot of the data in
column C1, and use these to comment on the assumption of Normality.
ii. In small samples, it is often difficult to check for Normality
by looking at a Histogram or a Box-and-Whisker plot. We often prefer to use a Normal Probability Plot.
Click on Graph Probability Plot
Single, and then click OK. Enter C1
in the Graph variables box and click
OK.
The graph you see is a Normal
Probability Plot, which plots theoretical probabilities from the Normal
distribution against observed probabilities from your data the diagonal line
represents equality between these two quantities, and so the closer the points
lie to this line, the more plausible the assumption of Normality. Comment on the assumption of Normality using
this plot.
40 |
42 |
49 |
31 |
43 |
52 |
50 |
45 |
44 |
39 |
33 |
34 |
36 |
35 |
42 |
Test the null hypothesis that there is no difference between the
average lead content of water in the two areas of
(a) What is the test statistics as calculated by Minitab? Does this match up with the test statistic
you calculate by hand for these data in tutorial?
(b) What is the exact p-value
for this hypothesis test? Does this
match up with the range for the p-value
you obtained by hand in tutorial?
Interpret this p-value and
form your conclusions.
(c) Minitab also gives the 95% confidence interval for the mean
difference. If this contains zero, then
this implies there is no significant difference between the two population
means. Interpret this confidence
interval does this match up with your conclusions in part (b)?
(d) Use Minitab to check the assumption that both samples are drawn
from a Normal distribution.
= £542 and
= £70.
Construct a 90% confidence interval for the average cost of collisions
of this kind by clicking on Stat Basic
Statistics 1-Sample t. Select Summarized data and enter the Sample size, Mean and Standard Deviation. Untick Perform hypothesis test, click Options, and change the Confidence level to 90.
Click OK twice. Save the output and compare this to the
interval you will calculate by hand in Tutorial 2, and repeat these steps to
calculate a 95% and 99% interval in Minitab.