You are on page 1of 8

1101 EXPERIMENTAL DESIGN: Homework #5 (Due: 1 Nov 2021)

Written by Christopher Andrew (M10901855)

Problems:

1. (80 pts) The tensile strength of portland cement is being studied. Four different mixing
techniques can be used economically. A completely randomized experiment was
conducted, and the following data collected:

Questions:
a. Test the hypothesis that mixing techniques affect the strength of the cement. Use α
= 0.05.
b. Construct a graphical display as described in “October 19, 2021 class” to compare
the mean tensile strengths for the four mixing techniques. What are your
conclusions?
c. Use the Fisher LSD method with α = 0.05 to make comparisons between pairs of
means.
d. Construct a normal probability plot of the residuals. What conclusion would you
draw about the validity of the normality assumption?
e. Plot the residuals versus the predicted tensile strength. Comment on the plot.
f. Prepare a scatter plot of the results to aid the interpretation of the results of this
experiment.
g. Rework part (b) using Tukey's test with a = 0.05. Do you get the same conclusions
from Tukey's test that you did from the graphical procedure and or the Fisher LSD
method?
h. Find a 95% confidence interval on the mean tensile strength of the portland cement
produced by each of the four mixing techniques. Also find a 95% confidence
interval on the difference in means for techniques 1 and 3. Does this aid you in
interpreting the results of the experiment?
Answers:
a. Null hypothesis is all groups’ mean are same.
Alternative hypothesis is all groups’ mean is not same
We can test the hypotheses by using the One-Way ANOVA to calculate the p-value.
The result of the calculation is as follows.

From the test, we get the p-value of 0.0004887, which is lower than the significant
level, thus we reject the null hypothesis and conclude that there is significant
affect to the strength of cement due to the differences on the mixing techniques
(mean of all groups are not same). Then after we know that there is significant
differences between the mixing technique effects, we have to check which one is
the most different by using either Fisher’s LSD or Tukey HSD tests. The differences
on each group can also be seen on the box plot as shown below.

b. The residual plot of the model is as follows:


c. Fisher’s LSD can be calculated by using the formula as follows.
Insert LSD formula
From the formula, we need to calculate each component, which is the Degree of
Freedom (DF) and Mean Square Within (MSW). These values can be calculated
using the function below.

And after executing it, we get the result as follows:

Thus, we can get DF = 12 and MSW = 12825.6875. With n or amount of data in


each sample is 4, we can calculate the LSD as follows:
The result is as follows:

d. The normal probability plot is as follows:


e. The residuals versus the predicted tensile strength plot are as follows (Residuals vs
Predicted Tensile Strength)

f. The scatter plot of the results of the mixing techniques is as follows: (Tensile
Strength VS Mixing Techniques)

g. Using Tukey’s HSD we can get a clearer result as it takes the number of groups into
consideration. Tukey’s HSD can be calculated by using the function below:

The result is as follows:


From the result above, we can see that there are no significant differences (p-value
> 0.05) between (MT1 x MT2), (MT1 x MT3), (MT2 x MT3) and significant
differences (p-value < 0.05) on the pair of (MT1 x MT4), (MT2 x MT4), (MT3 x
MT4). From this result, it can be concluded that the most significant different
compared to the other mixing techniques is MT4.
The result on LSD and HSD tests are different.
h. The 95% confidence interval on the means of all mixing techniques are as follows:

The 95% confidence interval on the means of the differences on technique 1 (MT1)
and technique 3 (MT3) can be calculated as follows

And the result is below:

There we can see that Mixing Technique 1 and Mixing Technique 3 has the
lowest differences, on the other hand, both MT1 and MT3 are the most similar
compared to the other MTs. Thus, for a larger scale of sample, both MT 1 and
MT3 can be combined to reduce the searching space.
2. (20 pts) Four catalysts that may affect the concentration of one component in a three-
component liquid mixture are being investigated. The following concentrations are
obtained from a completely randomized experiment:

Questions:
a. Do the four catalysts have the same effect on the concentration?
b. Analyse the residuals from this experiment.
Answers:
a. Null hypothesis is all groups’ mean are same.
Alternative hypothesis is all groups’ mean is not same
We can test the hypotheses by using the One-Way ANOVA to calculate the p-value.
The result of the calculation is as follows.

Because the p-value is less than significant level of 0.05, we reject the null
hypothesis and conclude that the four catalysts do not have the same effect on
the concentration. (all groups’ mean is not the same or there is significant
differences in the mean of the groups.)
b. The residual plots of the data is as follows:

.From the normal probability plot, we can see that the data is normally distributed
and it is able to be used for linear regression models.

You might also like