FACULTY OF BUSINESS MANAGEMENT

BACHELOR OF BUSINESS ADMINISTRATION (HONS) MARKETING

MGT555
BUSINESS ANALYTICS

ASSIGNMENT 2

Individually / Pair

STUDENT’S NAME:

NO NAME STUDENT ID CLASS

1. SHAFI HAIKAL BIN MOHD FAIROLSHAM 2023131249 BA2403A

LECTURER’S NAME:

MS. NOOR RITA MOHAMED KHAN

SUBMISSION: 14th JUNE 2024

QUESTION 1 - Determine the appropriate statistical test to use and formulate the hypotheses

a) As a dietitian, you want to see if the 2 weeks diet program offered to your 10 patients is
effective. You collected the data on your patients’ weights for their first visit and after 2 weeks.
The mean weight for the 1 st visit and 2 nd visit are 84.7kg and 81.5kg, respectively. Which
test would you conduct to determine whether the mean for the 1 st visit and 2 nd visit were
significantly different?
Test to use: Paired t-test. Since we are comparing the weights of the same patients before and after
the diet program, this is a paired (dependent) comparison.
Hypothesis:
The mean weight before and after the diet program is the same.
H0: μ1= μ2
H1:μ1=μ2

b) A health consultant claims that the average daily caloric intake of adults in a certain city is 2,200
calories. To verify this claim, a sample of 50 adults is selected, and their daily caloric intake is
recorded. The sample mean caloric intake is found to be 2,150 calories with a standard deviation
of 300 calories. Formulate the hypothesis to test the consultant’s claim and determine which test
to use.
Test to use: One t-test sample. The sample mean caloric intake is being compared to the
known population mean.

Hypothesis:
H0:μ=2200
H1:μ≠2200

c) FBM management would like to know if teaching MGT555 should be conducted via online or
face-to- face. To help FBM management in making this decision, we need to compare the
students’ performance among a sample of students who learned MGT555 via online and those
who learned via face-to-face. Formulate the hypothesis to test and determine which test to use.
Test to use: independent t-test. We are comparing the means of two independent groups (students
taught online vs. face-to-face).
Hypothesis:
H0: μonline=μface-to-face
H1:μonline=μface-to-face
 d) Before deciding on the advertising campaign for the new car model, a sample of 30 customers
were asked to choose between 1 st ad, 2 nd ad and 3 rd ad that most appealing to them. It also
wants to know whether the advertising campaign dependent on a particular age group (i.e.
among those aged under 25, 25-40, 41 and above). Formulate the hypothesis to test and
determine which test to use.
Test to use: Chi-square independent test. We are trying to find out if age group and advertisement
preference are related categorization variables.

Hypothesis:
H0: Ad preference is independent of age group
H1: Ad preference is dependent on age group

e) The palm oil farmers recorded the age of their palm oil trees, the amount of organic fertilizer
used, the amount of rainfall and their oil palm yield for the year of 2023. They want to know
whether the oil palm yield is affected by those conditions. Which test will be most appropriate
to use? Formulate the hypothesis for this test.
Test to use: multiple regression analysis. We want to see the effect of multiple variables (age of trees,
amount of organic fertilizer, amount of rainfall) on a dependent variable (oil palm yield).
Hypothesis:
H0:β1=β2=β3=0
H1:At least one βi=0
Alternative hypothesis: The oil palm yield is influenced by at least one of the independent variables.

f) A grocery chain wants to know if its three different types of advertisements (Radio, Banner,
Flyers) affect mean sales differently. They use each type of advertisement at its 3 different stores
for one month and measure total sales for each store at the end of the month. Which test will be
most appropriate to use? Formulate the hypothesis for this test.
Test to use: one-way ANOVA. We are comparing the means of sales across more than two groups
(three different types of advertisements).
Hypothesis:
H0: μRadio = μBanner = μFlyers 0
H1: At least one μi≠μj
QUESTION 2
A survey was conducted on 100 college students to determine their preference of food delivery apps.
The following data were obtained:

I. Using both critical value and p-value approaches, conduct a statistical test to determine if the
students’ choice of food delivery apps are related to their gender (using confident level of 90%)?
What the conclusion and implications of this finding?

Hypothesis
(H0): Gender and the applications used for food delivery are unrelated.
(H1): Gender and the apps that people use for food delivery are related.

Expected frequency for each cell is calculated using the formula:

(Row total)x(Column total)/Grand total

The chi-square statistic:

x^2: 6

The degrees of freedom (df) is calculated as:

(number of rows- 1) x (number of columns- 1)
(2-1) x (4-1)=3
P-Value : 0.094148967

With a 90% confidence level, the remaining 10% equals roughly 6.2514. Thus, 6.3889 >
6.2514 for x^2. Given its greater value, the null hypothesis is rejected.
QUESTION 3
The State of Ohio Department of Education has a mandated ninth-grade proficiency test that
covers mathematics, science, writing, reading and social studies. Below is the data of the Ohio
Education Performance Results in school districts of the great Cincinnati metropolitan area.

a) Calculate the mean scores for all the five tests (math, science, writing, reading and social
studies) for the districts of Indian Hill, Milford and Sycamore. Determine whether the mean
scores for these three districts are significantly different or not. You are to use both critical value
and p-value approaches.
Hypothesis:
H0: µ=µ0
H1: µ≠µ0

Sample Size : n = 31
Degrees of freedom : n-1 = 30
The significance level : α = 0.05
Tcritical : ±2.04

As a result, if the t values are within the rejection region, the null hypothesis will be rejected.

Indian Hill Milford Sycamore

Math 89 72 80
Science 91 76 84
Writing 95 82 85
Reading 98 86 88
Social 95 82 87

Hypothesis:
H0: µ1 = µ2 = µ3
H1: At least one of them is not equal
Anova test:
We apply ANOVA test

sample size of n=12

The significance level of α=0.05

 b) Test the hypothesis that the mean difference in math and science scores is zero. Use the
paired-sample procedure and use both the critical value and p-value approaches.

School District Math Science

Indian Hill 89 91
Sycamore 90 85
Milford 71 76

Hypothesis:
The mean difference between Maths and Science is zero.
H0: μ1-μ2 = 0
H1:μ1-μ2 ≠ 0

We apply the p-value and critical value approaches to t-test paired samples.
In the two-tailed test, we reject the null hypothesis because 0.03775 < 0.05, which is less than 0.05.
Based on the available data, this result suggests that there is enough evidence to draw the conclusion
that there is a significant difference between the math and science scores.

c) Examine the correlation matrix between math success (as dependent variable) and success
in the other subjects (as independent variables). From the correlation analysis, determine
the variables that are highly correlated with the success of math. Is multicollinearity a
problem?

The figure above displays the correlation result. The stronger the relationship between the variables, the
higher the correlation value. The social variable, with a correlation value of 0.857962, is highly
correlated with math success. Additionally, there is a multicollinearity issue with an absolute value
greater than 0.7 or 0.8. Because science and reading have a higher absolute value than math, it can
become a multicollinearity problem in math, as this table illustrates.

d) From the correlation analysis, further tests can be carried out to determine the subjects
that could significantly influence the success of math. Discuss how you would conduct
this test.

The results of the multiple regression analysis show that there is no correlation between the
independent variables—science, writing, reading, and social studies—and the dependent
variable, math.