0% found this document useful (0 votes)
9 views

Tutorial08 Solutions

Uploaded by

dmytro.pyrozhkov
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
9 views

Tutorial08 Solutions

Uploaded by

dmytro.pyrozhkov
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Tutorial 8

Question 1:
A pharmaceutical company is testing a new drug that they believe will lower the blood
pressure of patients who receive it. The company has conducted their clinical trial,
where they must compare their drug to that of a placebo. The placebo group receive
an inactive pill, whereas the treatment group receive the new drug. Each patient
belongs to one group and gets their blood measured once. The results from the trial
are as follows:
Placebo Treatment
Group Group
Mean 156.5 145.2
Standard Dev. 11.2 13.2
Sample Size 17 13
a. Use an F-test to test whether equal population variances can be assumed
b. Test whether the mean blood pressure differs significantly between the two
groups. Interpret the results.

Answers:
a. Step 1: Null and Alternative Hypothesis
H0: 𝜎!" = 𝜎#"
HA: 𝜎!" ≠ 𝜎#"
Two – tailed test
Step 2: Significance level
a = 0.05 2.89
Step 3: Calculate the test statistic 1.389
#!" $%.""
𝐹= ##"
= $$."" = 1.389
Step 4: Critical value from the F-table
𝐹'$ ,'" = 𝐹'! ,'# = 𝐹$%*$,$+*$ = 𝐹$",$, = 2.89
Step 5: Make conclusion
Since the test statistic, F = 1.389, falls in the ‘Fail to Reject’ region, we do not
have enough evidence to reject H0
Conclusion: There is insufficient evidence to suggest a difference in variance in
the two groups
b. Step 1: Null and Alternative Hypothesis
H0: No difference in blood pressure between groups
HA: There is a difference in blood pressure between groups
Two – tailed test
Step 2: Significance level
a = 0.05
Step 3: Calculate the test statistic (formulae from Lecture 13 slide 13)
(/$ *$)#$" 1(/" *$)#"" ($%*$)($%.")" 1($+*$)($$.")" "232.441"22+.25
𝑠-" = /$ 1/" *"
= $%1$+*"
= "4
= 146.35
777$7*6
6 777"7 $59."*$9,.9 $$.%
𝑡= = $'(.*+ $'(.*+
= 5.5, = 2.53
#"
% #"
% : 1
8 1 $* $,
&$ &"

Step 4: Critical value from the t-table


𝑡2.2"9,"4 = ± 2.048
Step 5: Make conclusion
Since the test statistic, t = 2.53, falls in the ‘Reject’ region, we have enough
evidence to reject H0
Conclusion: There is evidence to suggest a significant difference in blood
pressure between the groups. The evidence suggests that those in the
treatment group have a lower blood pressure, on average, indicating that the
treatment is better than the placebo.
Question 2:
To carry out an experiment, sometimes a catalyst is used. A catalyst increases the rate
of a chemical reaction without themselves being consumed in the experiment. For
some chemical process being conducted by Beckmann Coulter, they are analysing
two different catalysts and the affect they have on the mean yield. Catalyst 1 is
currently in use, however Catalyst 2 is cheaper and are willing to adopt this into the
experiment so long as it doesn’t affect the yield. They run the process 8 times using
Catalyst 1 and 8 times with Catalyst 2. The results of these 8 runs are given in the
table below.
Catalyst 1 Catalyst 2
Mean 92.26 mg 92.73 mg
Standard Dev. 2.39 mg 2.98 mg
a. Is there evidence that the variances differ between the two catalysts? For your
analysis, use a = 0.05.
b. Is there evidence that the mean yields differ between the two catalysts?
c. Based on your analysis, should Beckmann Coulter adopt Catalyst 2? Explain
your answer.

Answers:
a. Step 1: Null and Alternative Hypothesis
H0: 𝜎!" = 𝜎#"
HA: 𝜎!" ≠ 𝜎#"
Two – tailed test
Step 2: Significance level
a = 0.05 4.99
Step 3: Calculate the test statistic 1.55
#!" ".34"
𝐹= ##"
= ".%3" = 1.55
Step 4: Critical value from the F-table
𝐹'$ ,'" = 𝐹'! ,'# = 𝐹4*$,4*$ = 𝐹+,+ = 4.99
Step 5: Make conclusion
Since the test statistic, F = 1.55, falls in the ‘Fail to Reject’ region, we do not
have enough evidence to reject H0
Conclusion: There is insufficient evidence to suggest a difference in variance in
the two groups
b. Step 1: Null and Alternative Hypothesis
H0: Mean yields are the same (𝜇$ = 𝜇" )
HA: Mean yields are different (𝜇$ ≠ 𝜇" )
Two – tailed test
Step 2: Significance level
a = 0.05
Step 3: Calculate the test statistic (formulae from Lecture 13 slide 13)
(/$ *$)#$" 1(/" *$)#"" (4*$)(".34)" 1(4*$)(".%3)" $2".$9
𝑠-" = /$ 1/" *"
= 414*"
= $5
= 7.296
777$7*6
6 777"7 3".+%*3".", 2.5+
𝑡= = ,."-( ,."-(
= $.%9 = 0.35
# "
# " : 1
8 %1 % . .
&$ &"

Step 4: Critical value from the t-table


𝑡2.2"9,$5 = ± 2.145
Step 5: Make conclusion
Since the test statistic, t = 0.35, falls in the ‘Fail to Reject’ region, we do not
have enough evidence to reject H0
Conclusion: There is not enough evidence to suggest a significant difference
in the mean yields differ between the two catalysts.
c. Yes, there is no evidence suggesting that Catalyst 2 produces a mean yield that
differs from Catalyst 1. Since it is cheaper, this seems like a strategic move in
the current climate.
Question 3:
ODEON cinema in Castletroy want to test whether snack choices are linked to age
(adult or child). They conduct a survey with the results shown below.
Snack
Popcorn Nachos Total
Child 75 15 90
Age
Adult 20 40 60
Total 95 55 150
Using a Chi-Square test, investigate whether there is an association between age and
snack at ODEON cinema.

Answers:
Step 1: Null and Alternative Hypothesis
H0: There is no association between age and snack
HA: There is an association between age and snack
Step 2: Significance level
a = 0.05
Step 3: Calculate the test statistic
First calculate the expected value table
Snack
Popcorn Nachos
(90 x 95)/150 = (90 x 55)/150 =
Child
57 33
Age
(60 x 95)/150 = (60 x 55)/150 =
Adult
38 22
Chi-square assumption holds as E > 5
Test statistic
(;*<)" (+9*9+)" ($9*%%)" ("2*%4)" (52*"")" $4" $4" $4" $4"
𝜒" = ∑ <
= 9+
+ %%
+ %4
+ ""
= 9+
+ %%
+ %4
+ ""
=
%"5 %"5 %"5 %"5 4$22
9+
+ %%
+ %4
+ ""
= "23
= 38.76
"
Step 4: Critical value from the 𝜒 -table
"
𝜒2.29,$ = 3.841
Step 5: Make conclusion
Since the test statistic, 𝜒 " = 38.76, falls in the Reject’ region, we do have
enough evidence to reject H0
Conclusion: There is enough evidence to suggest an association between age
and snack preference at the ODEON cinema.

You might also like