EMILIO AGUINALDO COLLEGE

SCHOOL OF ENGINEERING AND TECHNOLOGY

QUANTITATIVE METHODS
(INCLUDING MODELING AND SIMULATION)

ASSESSMENT TASK 1
(Exploratory Data Analysis and Visualization)

Name (Student):
SHIEL DAWN AMON ALOJIPAN

Instructor / Professor:
ALEX HERNANDEZ

Submission Date:
12/02/2025
Objective:
The objective of this data visualization practice exercise is to analyze and interpret the iPhone
purchase dataset using various graphical representations. Through this exercise, students will:

● Gain hands-on experience with data visualization techniques in Python.

● Learn to use libraries such as Matplotlib and Seaborn for effective data representation.

● Identify patterns and trends in the dataset, including purchase behavior by age, gender, education
level, and location.
● Understand how different types of visualizations help convey meaningful insights from data.

Instructions:

● Please download the iphone purchase data set

(https://classroom.google.com/c/NzQ4ODU3NDAzNDMx/m/NzM3NjYwNjM2ODg5/details)
● Upload the data set under files of Google Colab.

● Use Google Colab to run the python code.

import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
# Load the dataset
df = pd.read_csv("iphone_purchases.csv")
# Convert Date of Purchase to datetime format
df["Date of Purchase"] = pd.to_datetime(df["Date of Purchase"])
# 1. Distribution of Age
df["Age"].plot(kind='hist', bins=20, edgecolor='black', title='Age Distribution')
plt.xlabel("Age")
plt.show()

Output

After observing the generated histogram, you might describe the findings like this:
"The histogram shows that the customer ages are mostly concentrated between [age range], with
a peak around [age]. The distribution appears to be slightly skewed to the [left/right], indicating a
higher proportion of [younger/older] customers. There are a few outliers above [age], representing
a small segment of older customers."
# 2. Count of Purchases by Gender
sns.countplot(x="Gender", data=df, palette="coolwarm")
plt.title("Purchases by Gender")
plt.show()

Output

After observing the generated countplot, you might describe the findings like this:
"The countplot reveals that [Gender] made significantly more purchases compared to [Gender],
 accounting for [Percentage]% of total purchases. This suggests a potential preference for
[product/brand] among [Gender] customers. The difference in purchase counts is substantial,
indicating a clear distinction in purchasing behavior between the two genders."

# 3. Count of Purchases by Education Level

sns.countplot(y="Education", data=df, palette="viridis")
plt.title("Purchases by Education Level")
plt.show()

Output
 "The countplot reveals that customers with [Education Level] have the highest number of
purchases, followed by [Education Level] and [Education Level]. This suggests a potential
correlation between education level and purchasing behavior for this product/service. There is a
noticeable trend of [increasing/decreasing] purchases with higher education levels."

# 4. Most Purchased iPhone Models

sns.countplot(y="Item Purchased", data=df, order=df["Item Purchased"].value_counts().index,
palette="pastel")
plt.title("Most Purchased iPhone Models")
plt.show()

Output

he iPhone 15 is the most purchased iPhone model, followed by [second most purchased model]
and [third most purchased model]. There is a noticeable difference in purchase frequency
 between the iPhone 15 and other models, indicating its strong popularity among customers. The
remaining iPhone models have relatively similar purchase counts, suggesting a more balanced
distribution for those models."

# 5. Total Sales by Location

df.groupby("Purchase Location")["Total Amount"].sum().sort_values().plot(kind='barh', color='skyblue',
title='Total Sales by Location')
plt.xlabel("Total Sales ($)")
plt.show()

Output
 The bar chart highlights that [Location 1] and [Location 2] are the top-performing locations,
generating the highest total sales and contributing [Percentage 1]% and [Percentage 2]% of overall
revenue, respectively. [Location 3] follows with a significantly lower sales figure, indicating a
potential need for further investigation or marketing efforts in that region. The remaining locations
show a relatively even distribution of sales, suggesting a balanced market presence across those
areas."

# 6. Purchase Trends Over Time

df.groupby("Date of Purchase")["Quantity"].sum().plot(figsize=(12,5), title='Purchase Trends Over Time',
color='purple')
plt.ylabel("Quantity Sold")
plt.show()
 Output

he plot reveals an overall [increasing/decreasing/stable] trend in purchase quantity over time.

There appears to be a seasonal pattern with sales peaking around [month/quarter/year] and
declining during [month/quarter/year]. A significant increase in sales occurred in [month/year], as
indicated by a sharp rise in the percentage change line. This could be attributed to [potential
reason, e.g., new product launch, marketing campaign]. Overall, the purchase trends show a
[positive/negative/mixed] outlook for sales performance.

# 7. Boxplot of Total Amount by Gender

sns.boxplot(x="Gender", y="Total Amount", data=df, palette="coolwarm")
plt.title("Total Amount Spent by Gender")
plt.show()
 Output

The boxplot reveals that [Gender] tend to spend more on average, as indicated by a higher median
spending amount. The spread of spending is wider for [Gender], suggesting greater variability in
their purchase amounts. There are a few outliers for [Gender], representing unusually high
spending instances. Overall, [Gender] contribute a larger percentage ([Percentage]%) to the total
spending compared to [Gender] ([Percentage]%)."

# 8. Correlation Heatmap
plt.figure(figsize=(8,6))
sns.heatmap(df.corr(numeric_only=True), annot=True, cmap="coolwarm", linewidths=0.5)
plt.title("Correlation Heatmap")
plt.show()

Output
 The correlation heatmap reveals a strong positive correlation between [Variable 1] and [Variable
2], suggesting that they tend to move together in the same direction. There is a moderate negative
correlation between [Variable 3] and [Variable 4], indicating an inverse relationship. [Variable 5]
shows weak or no correlation with other variables. These findings provide insights into the
relationships between different features in the dataset and can guide further analysis or
modeling."

# 9. Please provide the mean, median, mode of the purchase data.

Output