ASSIGNMENT 2

School of Business

MSMG-5100: Applications of Statistics

Submission of Assignment 2

Member Student ID E-mail address

Md Arif Khan 0765971 [email protected]
Abhishek Rudra 0776509 [email protected]
Arif Bin Hossain 0759031 [email protected]
Fahim Alam 0573669 [email protected]
Kazi Anwar Hosain 0768362 [email protected]
Mohammad Asif Bin Zahir 0783630 [email protected]
Ashik Anwar Deep 0768501 [email protected]

Submission Date: March 31, 2023

ASSIGNMENT 2

Answer to question 1

A. Total services provided (n) = 100,000

The researcher is interested in knowing the fraction of customers who are satisfied or very satisfied.
So, the population parameter is the proportion of persons who are satisfied or very satisfied by the
service.

B. The sample proportion is the population proportion parameter's point estimate.

So, the sample proportion of persons who are satisfied or very satisfied
= (no. of persons who are satisfied or very satisfied) / (Total number of persons in the sample)
= (360 / 400)
= 0.90
Therefore, the value of point estimate of the population proportion = 0.90
C.
CI=p ± zsqrt((p(1-p))/n).

where:
CI=confidence interval p=proportion of customers pleased or extremely satisfied
z = z-score for desired confidence level (95%).
n=sample size

Given:
360 consumers were happy or extremely satisfied.

Thus:
p̂ = 360/400 = 0.9
A simple normal distribution calculator or table provides 1.96 for a 95% confidence level z-score.

Values yield:

CI = 0.9 ± 1.96sqrt((0.9(1-0.9))/400) = 0.038 (0.862, 0.938)

ASSIGNMENT 2

We are 95% positive that 86.2%–93.8% of consumers are "satisfied" or "very satisfied" with the
service. If we repeated this survey several times, 95% of the time, the genuine proportion of
satisfied customers would fall within this range. Making decisions and getting better at this time
period. It implies that the company will have lots of happy customers. The calculated proportion
can be higher or lower than the interval due to sample variability.
D. To construct a 99% confidence interval, we can follow the same formula but use a different
critical value. For a 99% confidence level, the critical value (z) is 2.5.
To construct a 99% confidence interval, we have to use the same formula, but use a different
critical value. The critical value (z) is 2.5 for 99% confidence level.
After putting the values, we will get:
P = 0.9 ± [2.576 √ ((0.9(1-0.9))/400) ] = 0.9 ± 0.03864
Hence, [0.86136, 0.93864] is the figure we get for the 99% confidence interval. The meaning of
this is that we are 99% certain that the true percentage of all the firm's customers who are either
"satisfied" or "extremely satisfied" is between 86.14% and 93.86%.
E. To determine the necessary sample size to achieve a 99% confidence interval no more than 3
percentage points wide, we can use the formula:
We can use the formula below to figure out the required size of sample to achieve 99% confidence
interval that is not greater that 3% points wide.
n = (z^2 p^' q^')/e^2
In this equation, p' and q' are the estimated values for the population parameter, and p' = 1- q'. The
Sampling error (e) is 0.03 in this instance, and the z-value correlates to the desired confidence
level and in this example, 2.576 for 99%. As (0.5) (0.5) is the greatest possible product of p' and
q', we use 0.5 for both in this case. To be 99% sure that we are 3 percentage points wide, the largest
product gives us the highest sample size (n).
After putting in the values, we get:
n = (〖2.576〗^2 (0.5) (0.5))/〖0.03〗^2 = 1843.27
Therefore, to achieve a 99% confidence interval which is no more than 3% points wide,
approximately 1843 customers are required.
ASSIGNMENT 2

Answer to question 2

We need to conduct a hypothesis test to determine whether the new sterilization guidelines have
reduced the proportion of patients who experience an HAI.

Let p1 be the proportion of patients who experience an HAI before the implementation of the new
guidelines, and p2 be the proportion of patients who experience an HAI after the implementation
of the new guidelines.

The null hypothesis (H0) is that there is no difference between the proportions of patients who
experience an HAI before and after the implementation of the new guidelines.
That is, H0: p1 = p2

The alternative hypothesis (Ha) is that the proportion of patients who experience an HAI after
implementing the new guidelines is lower than the proportion of patients who experience an HAI
before implementing the new guidelines.
That is, Ha: p2 < p1

We can use the z-test for two proportions to conduct the hypothesis test. The test statistic is
calculated as:
z = (p1 - p2) / sqrt(p * (1 - p) * (1/n1 + 1/n2))

where p = (x1 + x2) / (n1 + n2), x1 and x2 are the number of patients who experienced an HAI
before and after the implementation of the new guidelines, respectively, and n1 and n2 are the
sample sizes before and after the implementation of the new guidelines, respectively.

Using the given information, we can calculate:

p1 = 75/480 = 0.15625 p2 = 67/460 = 0.14565 n1 = 480 n2 = 460
p = (75 + 67) / (480 + 460) = 0.15163

The test statistic is:

z = (0.15625 - 0.14565) / sqrt(0.15163 * (1 - 0.15163) * (1/480 + 1/460)) = 1.379
ASSIGNMENT 2

Using a standard normal distribution table, the critical value for a one-tailed test at a 5% level of
significance is -1.645.

Since the calculated z-value of 1.379 is greater than the critical value of -1.645, we fail to reject
the null hypothesis. Therefore, there is insufficient evidence to conclude that the new sterilization
guidelines have reduced the proportion of patients who experience an HAI.
ASSIGNMENT 2

Answer to question 3

We need to run a hypothesis test to determine whether we should advise customers to purchase
tools from the alternative provider. The null hypothesis (H0) states that the average usable lifespan
of the tools from the alternate supplier is the same as the average usable lifespan of the existing
tools (900 hours). The alternative hypothesis (Ha) states that the tools from the alternate supplier
have a mean usable lifetime that is different from 900 hours.

The mean useful lifetime of the existing tools and the tools from the alternative supplier can be
compared using a two-sample t-test. To contrast the sample mean with the known population mean
of 900 hours, we can apply a one-sample t-test for the other supplier's tools as we only have one
sample.
calculating the sample mean and standard deviation of the alternate supplier's tools:
Sample mean = (880 + 930 + 920 + 905 + 925 + 890) / 6 = 907.5 hours Sample standard deviation
= sqrt(((880-907.5)^2 + (930-907.5)^2 + (920-907.5)^2 + (905-907.5)^2 + (925-907.5)^2 + (890-
907.5)^2)/5) = 17.05 hours
The t-statistic can be calculated as:
t = (sample mean - population mean) / (sample standard deviation / sqrt(sample size))
t = (907.5 - 900) / (17.05 / sqrt(6))
t = 1.49
Using a t-distribution table with degrees of freedom (df) = n-1 = 5, we can get the crucial t-value
for a two-tailed test at a 95% confidence level. About 2.571 is the essential t-value.
We cannot rule out the null hypothesis since the calculated t-statistic (1.49), which is below the
crucial t-value (2.571), is not statistically significant. This indicates that, with a 95% confidence
level, it is not possible to conclude that the average usable lifetime of the tools from the alternative
supplier differs from the average useable lifetime of the existing tools.

Thus, we are unable to suggest purchasing tools from the alternative provider based on the
statistical analysis.
ASSIGNMENT 2

Answer to question 4

We can use an analysis of variance (ANOVA) test to determine whether the mean outputs of the
three plants are equal. The ANOVA test is a statistical technique used to determine whether there
are any significant differences between the means of two or more groups.

In this instance, there are three groups: Plants 1, 2, and 3. We want to determine whether the means
of these groups are equal. The three groups' means being equal is the null hypothesis for the
ANOVA test, whereas at least one group's mean being different is the alternative hypothesis.
The sum of squares between groups (SSG) and the sum of squares within groups (SSW) are
computed before the ANOVA test is run. The SSG quantifies the variation in the three groups'
means, whereas the SSW quantifies the variation within each group.

We can determine the SSG and SSW by using the information provided.

SSG .

= [(240 + 265 + 270 + 235 + 250 + 280)/6 – (262.5)]^2 x 6

+ [(260 + 270 + 275 + 290 + 280 + 290)/6 – (272.5)]^2 x 6

+ [(280 + 280 + 300 + 250 + 280 + 300)/6 – (280)]^2 x 6

= 1825

SSW .

= [(240 - 262.5)^2 + (265 - 262.5)^2 + (270 - 262.5)^2 + (235 - 272.5)^2 + (250 - 272.5)^2 + (280
- 272.5)^2]

+ [(260 - 272.5)^2 + (270 - 272.5)^2 + (275 - 272.5)^2 + (290 - 272.5)^2 + (280 - 272.5)^2 + (290
- 272.5)^2]
ASSIGNMENT 2

+ [(280 - 280)^2 + (280 - 280)^2 + (300 - 280)^2 + (250 - 280)^2 + (280 - 280)^2 + (300 - 280)^2]
= 2370

The degrees of freedom (df) for both the SSG and SSW must be calculated next. While the df for
the SSW is n-k, where n is the total sample size (in this case, n=18), the df for the SSG is k-1,
where k is the number of groups (k=3 in this case).
dfSG=3-1=2, and dfSW=18-3=15.
Now, by dividing the sum of squares by the corresponding degrees of freedom, we can compute
the mean square (MS) for both the SSG and SSW.
MSG equals SSG/dfSG.
= 1825/2 .
= 912.5 .
SSW divided by dfSW equals MSW.
= 2370/15 .
= 158.
By dividing the MSG by the MSW, we can finally determine the F-statistic.
F equals MSG/MSW.
= 912.5/158 .
= 5.77.
We must contrast this F-statistic with the F-distribution with dfSG=2 and dfSW=15 in order to
determine whether it is significant at the 5% level of significance. We can determine the critical
value of F for a significance level of 5% and these degrees of freedom to be 3 using a table of the
F-distribution or statistical software.