UNIVERSITI TEKNOLOGI MARA KAMPUS PUNCAK PERDANA

SCHOOL OF INFORMATION SCIENCE

INFORMATION ANALYTICS
(IMC501)
ASSIGNMENT 2 (40%)
COMPERENHENSIVE DATA ANALYSIS USING SAS
( OmegaKu Sdn Bhd )

PREPARED BY:

NAME STUDENT ID

HUMAIRA BINTI ABD HADI 2025121429

MUHAMMAD NAJWAN BIN MOHD MAZAN 2025148927

AMIR AJMAL BIN HASSAN 2025117295

ADAM BIN MOHD LATIFF 2025178371

GROUP:
CDIM2623C

PREPARED FOR:
MISS ALIYA BINTI RAMLI

SUBMISSION DATE: WEEK 15

ACKNOWLEDGEMENT

First and foremost, we would like to praise and thank Allah SWT for giving us strength
and because of His blessing, we finally managed to accomplish this assignment. Without His
blessing, we wouldn’t have gone this far. This assignment cannot be completed without effort
and co-operation from our group members, which consist of Najwan, Humaira, Adam and
Amir. We always work hard to produce a good task with our full commitment and responsibility.

Therefore, we would like to acknowledge our lecturer Miss Aliya because without her
guidance our assignment cannot be done properly like this. She always gives us support and
guidance on how to do our assignment to produce a good outcome. She inspired us greatly
to work on this project. We also like to thank her for teaching us in this course. She always
gives us support and guidance on how to do our assignment to produce a good outcome. She
inspired us greatly to work on this project. We also like to thank her for teaching us in this
course.

Finally, we were extremely grateful to our parents for their love, prayers, caring, and
sacrifices for educating and preparing us for our future. Their encouragement when the times
got rough are much appreciated and noted. It was a great comfort and relief to know that they
are always by our side no matter what happens

1
 TABLE OF CONTENTS

ACKNOWLEDGEMENT ....................................................................................................... 1
LIST OF FIGURES ............................................................................................................... 3
1.0 INTRODUCTION............................................................................................................. 4
2.0 DESCRIPTIVE STATISTICS ........................................................................................... 5
2.1 Summary Statistics ................................................................................................... 5
2.2 Discussion of Trends and Insights........................................................................... 5
2.3 Data Visualizations .................................................................................................... 6
3.0 INFERENTIAL STATISTICS ........................................................................................... 9
3.1 T-test Analysis ........................................................................................................... 9
3.2 ANOVA Analysis ...................................................................................................... 11
3.3 Regression Analysis ............................................................................................... 14
4.0 CONCLUSION .............................................................................................................. 16
5.0 RECOMMENDATIONS ................................................................................................. 17
6.0 APPENDICES ............................................................................................................... 18

2
 LIST OF FIGURES

NO Content Page
1 Figure 1: Histogram of Client Age 6
2 Figure 2: Bar Chart of Number of Transactions per Product Category 6
3 Figure 3: Box Plot of Total Amount Spent by Gender 7
4 Figure 4: Scatter Plot of Price per Unit vs Quantity Purchased 7
5 Figure 5: Line Chart of Average Satisfaction Rating by Store City 8
6 Figure 6: Comparison of total amount between female and male client 9
7 Figure 7: Plots of total amount 9
8 Figure 8: Satisfaction rating between clients by P002 and P003 10
9 Figure 9: Plots of satisfaction rating 10
10 Figures 10: ANOVA Analysis of total amount and payment method 11
11 Figures 11: Comparison of the mean total satisfaction ratings and 12
promotional codes
12 Figures 12: Comparison of product categories bought according to the 13
client's age
13 Figures 13: Simple Linear Regression 14
14 Figures 14: Multiple Linear Regression 15

3
1.0 INTRODUCTION

In the age of data-driven decisions, analyzing and interpreting data is an essential skill for
organizations aiming to improve their operational efficiency and customer satisfaction. This
task emphasizes utilizing different information analytics methods on a practical dataset
supplied by OmegaKu Sdn Bhd, an online retailer located in Kuala Lumpur. This analysis aims
to identify patterns, trends, and relationships in customer buying behavior using descriptive
and inferential statistical techniques.

The dataset comprises 200 entries featuring comprehensive details on customer

transactions, encompassing product specifications, customer demographics, store data, and
promotional efforts. Important factors consist of Product_Category, Total_Amount,
Client_Gender, Payment_Method, Promotion_Code, and Satisfaction_Rating. This report
seeks to derive meaningful insights from these variables analyzed through SAS, which can
aid in strategic business decision-making.

This evaluation is performed with the SAS statistical software and encompasses five key
elements: summary statistics, data visualizations, t-tests, ANOVA analyses, and regression
modeling. This thorough approach allows for a better comprehension of buying behaviors,
customer contentment, and possible elements that affect sales performance at OmegaKu Sdn
Bhd.

4
2.0 DESCRIPTIVE STATISTICS

2.1 Summary Statistics

Standard
Variable Mean Median Minimum Maximum
Deviation
Quantity
2.04 2.00 1.49 1.00 8.00
Purchased
Price per Unit (RM) 177.55 77.00 265.87 15.00 1359.90
Total Amount (RM) 341.77 119.80 559.87 15.00 4079.70
Client Age (years) 37.22 35.50 13.10 20.00 67.00
Satisfaction Rating 3.56 4.00 1.13 1.00 5.00

2.2 Discussion of Trends and Insights

1. Quantity Purchased
• Mean = 2.04, Median = 2.0, indicating a slight right skew.
• Most customers purchase around 2 items per transaction.
• The standard deviation is low, showing consistent purchase behaviour.

2. Price per Unit

• Mean = RM 177.55, but Median = RM 77.00 — a big gap suggests right-change
distribution (some very expensive items raise the average).
• Max = RM 1359.90 shows the presence of high-priced items, likely in electronics.

3. Total Amount
• Mean = RM 341.77, Median = RM 119.80 — again, change by high-value
purchases.
• High standard deviation (RM 559.87) reflects wide variability in transaction totals.
4. Client Age
• Mean = 37.22 years, Median = 35.5 years — symmetric and normally distributed.
• Age range is from 20 to 67 indicating a broad customer base.

5. Satisfaction Rating
• Mean = 3.56, Median = 4.0, with a max of 5.0 — customers tend to rate experiences
positively.
• Some ratings are as low as 1.0, suggesting a few outliers of dissatisfaction.

5
2.3 Data Visualizations
The analysis of the dataset using various visualizations reveals several key insights.

(Figure 1: Histogram of Client Age)

Firstly, the histogram of client age indicates that most customers fall within the age range of
25 to 45 years old. This suggests that young and middle-aged individuals are the company's
primary target market, and marketing initiatives should target them to increase engagement
and sales.

(Figure 2: Bar Chart of Number of Transactions per Product Category)

The bar chart displaying the number of transactions by product category shows that certain
categories, such as Electronics or Groceries, are significantly more popular than others. This

6
highlights the product categories that generate the highest sales and, as a result, ought to be
given more consideration in strategic business choices, inventory planning, and promotions.

(Figure 3: Box Plot of Total Amount Spent by Gender)

The box plot comparing total spending by gender reveals that while both male and female
customers have similar median spending levels, male customers show a wider range in their
spending and include more high-value purchases. This implies that male customers could be
more likely to spend more money or make larger purchases, which might be used for loyalty
plans or targeted advertising.

(Figure 4: Scatter Plot of Price per Unit vs Quantity Purchased)

7
The scatter plot analyzing the relationship between price per unit and quantity purchased
shows a clear negative trend. As the price of a product increases, the number of units
purchased generally decreases. This behaviour indicates that customers are price-sensitive,
meaning that higher prices may discourage bulk buying. Strategies like bundles, discounts, or
reduced packaging might work well to boost sales of more expensive goods.

(Figure 5: Line Chart of Average Satisfaction Rating by Store City)

Lastly, the line chart showing average customer satisfaction ratings by store city reveals
notable differences in customer experiences across locations. Some cities consistently score
higher in satisfaction, while others fall below average. This suggests that operational or
service-related issues may exist in certain cities, and these areas should be investigated
further to identify and address possible causes. High-performing cities could serve as
benchmarks for best practices.

In summary, the data reveals important trends regarding customer demographics, purchasing
behaviour, product popularity, price sensitivity, and geographic performance. These insights
can guide data-driven decisions in marketing, pricing, customer service, and inventory
management to enhance overall business performance.

8
3.0 INFERENTIAL STATISTICS
3.1 T-test Analysis

(Figure 6: Comparison of total amount between female and male client)

Comparisons of the Total_Amount spent between female and male clients. We conducted and
an independent samples t-test. Average Total_Amount for Male Customers (M = RM455. 3,
SD = RM680. 5), which is significantly more than female clients (M = RM211. 2, SD = RM335.
1). This difference was also statistically significant, t (159.13) = -3.28, p =.0013, which is below
the significance level of 0.05.

(Figure 7: Plots of total amount)

Hence null hypothesis is rejected because there is no significant difference in the mean total
amount spent between male and female clients. Last but not least, we conclude that client
gender does effect on total spending at OmegaKu Sdn Bhd.

9
 (Figure 8: Satisfaction rating between clients by P002 and P003)

Independent samples t-test was used to compare satisfaction rating among customers
serviced by P002 and P003. The average satisfaction of P002 was 3.5652 (SD = 1.1211) and
that of P003 was 3.5333 (SD = 0.7432). The mean difference was small (0.0319) and the t-
test was non-significant, t (35.972) = 0.11, p = 0.9166. As p-value > 0.05, thus we do not reject
null hypothesis.

(Figure 9: Plots of satisfaction rating)

This means there is no significant difference in the rating of satisfaction of clients who are
being handled by Promoter P002 against Promoter P003. Hence, promoter seems not to have
a substantial effect on client satisfaction in this group.

10
3.2 ANOVA Analysis
One-way ANOVA Analysis of the effects of different payment methods and each method's
purchase amount.

(Figures 10: ANOVA Analysis of total amount and payment method)

To find out if the type of payment method has a significant impact on the Total_Amount spent
by clients, a one-way ANOVA was performed. The findings indicated that there was no
statistically significant difference in the means (F(2, 166) = 0.50, p = 0.6051).

We are unable to determine out the null hypothesis since the p-value is higher than 0.05. This
suggests that customers who paid with cash, credit cards, or e-wallets were not significantly
different in their typical purchase amounts.

Consequently, at OmegaKu Sdn Bhd, the choice of payment method has little impact on the
total purchase cost. For alternative hypothesis, at least one payment method has a
significantly different mean Total_Amount.

11
One-way ANOVA Analysis of the comparison of the mean total satisfaction ratings and
promotional codes.

(Figures 11: Comparison of the mean total satisfaction ratings and promotional codes)

To examine whether satisfaction ratings differed significantly across 15 different promotion

codes a one-way ANOVA was conducted. The analysis revealed that the differences in
Satisfaction_Rating were not statistically significant, F(14, 185) = 0.79, p = 0.6769.

We are unable to determine out the null hypothesis since the p-value is higher than 0.05. This
suggests that type of promotion used by clients does not have a significant effect on their
satisfaction levels.

Thus, we conclude that client satisfaction is not significantly influenced by the promotion code
used at OmegaKu Sdn Bhd. For alternative hypothesis, at least one promotion code results in
a significantly different satisfaction rating.

12
One-way ANOVA Analysis of the comparison of the mean of product categories bought
according to the client's age.

(Figures 12: Comparison of product categories bought according to the client's age)

To determine whether different client age groups purchase different types of product
categories a one-way ANOVA was conducted. The results were not statistically significant,
F(3, 196) = 1.39, p = 0.2460.

We fail to reject the null hypothesis, since the p-value is greater than 0.05. This means there
is no statistically significant difference in the average product category purchased across age
groups.

Therefore, the age of clients does not appear to influence the type of product category bought
at OmegaKu Sdn Bhd. For alternative hypothesis, at least one age group purchases a
significantly different product category compared to the others.

13
3.3 Regression Analysis
Simple Linear Regression

(Figures 13: Simple Linear Regression)

To figure out if Satisfaction_Rating substantially predicts Total_Amount spent by clients, a
simple linear regression analysis was performed.

With an R-square value of 0.0057 and F(1, 198) = 1.13, p = 0.2892, the model was not
statistically significant. This suggests that fewer than 1% of the variation in Total_Amount can
be explained by Satisfaction_Rating.

The Satisfaction_Rating regression coefficient was β = 37.24 (p = 0.2892), indicating that the
Total_Amount should rise by RM37.24 for every additional point of satisfaction. However, this
impact is not statistically significant.

We are unable to rule out the null hypothesis since the p-value is higher than 0.05. As a result,
Total_Amount spent by customers at OmegaKu Sdn Bhd is not substantially predicted by
Satisfaction_Rating. For alternative hypothesis, satisfaction rating significantly predicts total
amount.

14
Multiple Linear Regression

(Figures 14: Multiple Linear Regression)

To find out if Client Age, Satisfaction Rating, and Product Code substantially predict
Total_Amount spent, a multiple linear regression analysis was performed. With an R-squared
value of 0.0614 and F(3, 196) = 4.28, p = 0.0060, the entire model was statistically significant.
This suggests that the three predictors account for around 6.14% of the variation in
Total_Amount.

The considerable positive effect of Client_Age (β = 6.12, p = 0.0413) indicates that older clients
often spend a little more.

A substantial positive effect was also seen for Product_Code (β = 109.06, p = 0.0034),
indicating that some product categories are linked to increased expenditure. A significant
predictor was not Satisfaction_Rating (β = 36.73, p = 0.2844). We reject the null hypothesis
because the model contains significant variables and is statistically significant. This implies
that customers' spending patterns at OmegaKu Sdn Bhd are influenced by both product
category and age. For alternative hypothesis, at least one predictor significantly affects total
amount.

15
4.0 CONCLUSION

Finally, the examination of client transactions at OmegaKu Sdn Bhd yielded some
significant discoveries. The bulk of the company's consumers are between the ages of 25 and
45, indicating that young to middle-aged individuals are its primary target market. Most clients
buy about two things per visit, with a preference for modestly priced products, however a small
number of high-priced items dramatically raises the average price per unit. In terms of product
trends, electronics and groceries appear to be the most popular, accounting for a significant
portion of total sales. Furthermore, the data reveals a definite negative link between product
price and quantity purchased, demonstrating that customers are price sensitive and less
inclined to purchase bigger quantities of more expensive things.

The data also found gender-based spending patterns, with male customers spending
more than female customers, which was statistically significant. Overall, customer satisfaction
was high, but it had no substantial impact on total spending. Similarly, promotional codes and
promoter assignment had little or no impact on consumer happiness. Notably, customer
satisfaction scores changed depending on the store's location, indicating that service quality
and operational efficiency vary by city. Regression analysis revealed that customer age and
product category had a substantial impact on overall expenditure, however satisfaction ratings
are not strong predictors of purchase amounts.

16
5.0 RECOMMENDATIONS

Based on these findings, several recommendations can be made to help OmegaKu Sdn
Bhd improve its business performance. The company should focus its marketing efforts on
customers aged between 25 and 45 years, as this group forms the largest segment of its
clientele. Offering bundle deals or discounts on higher-priced items could help address price
sensitivity and encourage more purchases of expensive products. Inventory and promotional
strategies should prioritize high-performing product categories, particularly electronics and
groceries, to maximize sales.
Since male customers tend to spend more, the company could consider targeted loyalty
programs or promotions specifically designed to attract and retain this segment. Moreover, it
is recommended that OmegaKu Sdn Bhd investigate cities with lower satisfaction ratings to
identify service or operational issues, while using high-performing locations as models for best
practices.
Age-based promotions and category-focused sales strategies should also be developed,
as age and product category have been identified as significant predictors of customer
spending.
Lastly, since current promotion codes do not seem to effectively influence satisfaction or
sales, the company should review and redesign its promotional offers to ensure they provide
real value to customers and encourage higher spending.

17
6.0 APPENDICES

18
19
20
21
22
23
24
25