FINAL ASSESMENT

SEMESTER 2 SESI 2019/2020

SBS 3013 (Group A)

CASE STUDY 1

No. Name Matrix No.

Date of Submission : 21ST August 2020

CASE STUDY 1
Question
StopSmoking is a product designed to help people quit smoking and no other products have to be
more effective. StopSmoking provides significant advantages compare to all other products. For
instance, StopSmoking is simpler to use. The product is also significantly less expensive than the
other leading brands. The effectiveness of StopSmoking product has been tested in two
independent experiments as shown in Table 1.
Table 1

Quit rates
StopSmoking Brand A
Experiment 1

17% 13%
StopSmoking Brand B
Experiment 2

18% 15%

How do you interpret the results and information given above from the statistical context?
You would like to conduct another experiment on your latest product in comparison to the
StopSmoking product. How would you determine the significant level of your statistical
analysis? Describe on how would your decision affect the type I error and type II error.

Answer
Interpretation According to Table 1
Two different experiment, 1 and 2 was conducted to show quit rates of smokers using Stop
Smoking, a product designed to help people quit smoking and no other products have to be more
effective. Experiment 1, shows that 17% of users of Stop Smoking product was able to quit,
compared to Brand A, 13% of users quit smoking. Experiment 2, shows that 18% of users of
Stop Smoking product was able to quit, compared to Brand B, 15% of user quit smoking. The
quit rate between experiment 1 and 2 is almost the same, the difference in the percentage could
be due to the amount of people selected to try the product. Different in the number of users will
give a different rate. If compared between Brand A and Brand B, it seems Brand B is more
 CASE STUDY 1 |3

effective as 15% has quit smoking using Brand B. The preferred product is still Stop Smoking as
it is easier to use and much cheaper and no side effects to the users.
Conducting another Experiment on Your Latest Product in Comparison to the Stop
Smoking Product

If another experiment was to be conducted on the latest product in comparison to the Stop
Smoking product, a proper sample size must be selected. An equal sample size will give a better
perspective of the product. Besides that, smokers finding ways to quit will always look for a
cheaper, saver and ideal product as it can be a long-term process. The advantages must be
attractive. A design to test the mean rate of quitter to that of the other product. A confidence
level of 95% and mean between two different sample data can be calculated to test null and
alternative hypothesis, so at least a 5% significance level should be taken.

Significant Level of Your Statistical Analysis

i. The general p is more or less than 0.05.

What is Type I and Type II Error?

A type 1 error is otherwise called a false positive and happens when an analyst mistakenly rejects
a genuine invalid speculation. This implies your report that your findings are significant
whenever in truth they have happened by some coincidence.

The likelihood of making type I error is spoken to by your alpha level (α), which is the p-value
below which you reject the invalid speculation. A p-value of 0.05 demonstrates that you are
eager to acknowledge a 5% chance that you are incorrect when you reject the null hypothesis.

You can lessen your risk of submitting a type I error by utilizing a lower value for p. For
instance, a p-value of 0.01 would mean there is a 1% possibility of submitting a Type I error.
 CASE STUDY 1 |4

Nonetheless, utilizing a lower value for alpha implies that you will be more averse to recognize a
genuine difference if one really exists (thus risking a type II error).

What is Type II Error?

A type II error is otherwise called a false negative and happens when an analyst fails to reject an
invalid theory which is extremely false. Here an analyst concludes there is not a significant
effect, when really there truly is.

The probability of making a type II error is called Beta (β), and this is identified with the
intensity of the statistical test (power = 1 - β). You can decrease your risk of committing a type II
error by ensuring your test has enough power.

You can do this by guaranteeing your example size is sufficiently enormous to recognize a down
to earth contrast when one really exists.

To Describe on How Would the Decision Affect the Type I Error and Type II Error

Type 1 Error: Reject the other product will be better, and accept Stop Smoking and which can
be perceived as the product produced better results, but in reality, the latest product rates of
quitting were better.

Type 2 Error: Accept that the Stop Smoking product is better, as it was out null hypothesis and
in reality, latest product was better.
 CASE STUDY 1 |5

References

[1] Hajek P, Phillips-Waller A, Przulj D, et al. A randomized trial of e-cigarettes versus

[2] CDC. Adult smoking cessation–the use of e-cigarettes. CDC website.

[3] CDC. How to quit smoking. CDC website. gov/tobacco/campaign/tips/quit-