Understanding Descriptive vs Inferential Statistics: A Complete Guide for Beginners

📗 Chapter 5: Practical Applications and Tools

Apply Statistics to Real-World Problems Using Python, Excel, and Visualization Libraries

🧠 Introduction

Up to this point, you’ve learned the theory behind descriptive and inferential statistics. Now it’s time to put those concepts into action using real-world tools and use cases.

In this chapter, we’ll explore how statistics power decisions in business, healthcare, science, and tech — and how you can use tools like Python, Excel, and visualization libraries to bring your insights to life.

You’ll learn:

• How to structure and analyze data with descriptive and inferential statistics
• Tools used by analysts and data scientists
• Project examples with code samples
• Practical workflows using Python, Excel, and BI dashboards

📘 Section 1: Practical Scenarios for Statistical Analysis

📘 Section 2: Tools for Statistical Analysis

📘 Section 3: Using Excel for Descriptive & Inferential Stats

Descriptive in Excel:

Inferential in Excel (with Data Analysis ToolPak):

1. Enable Data Analysis Toolpak
2. Go to Data > Data Analysis > t-Test, ANOVA, Correlation

📌 Example: T-Test in Excel

Compare mean satisfaction scores between two branches.

Use t-Test: Two-Sample Assuming Unequal Variances in Data Analysis ToolPak.

📘 Section 4: Using Python for Applied Statistics

📌 Dataset: Titanic

python

import pandas as pd

df = pd.read_csv("https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv")

Descriptive Summary:

python

df.describe()

df['Age'].mean(), df['Age'].median(), df['Age'].mode()[0]

Inferential Analysis: Survival vs. Gender

python

import scipy.stats as stats

female = df[df['Sex'] == 'female']['Survived']

male = df[df['Sex'] == 'male']['Survived']

t_stat, p_val = stats.ttest_ind(female, male)

print("t-statistic:", t_stat, "p-value:", p_val)

📊 Interpretation:

If p < 0.05 → Survival rate difference between genders is statistically significant.

Visualization with Seaborn:

python

import seaborn as sns

import matplotlib.pyplot as plt

sns.boxplot(x='Sex', y='Age', data=df)

plt.title("Age Distribution by Gender")

plt.show()

📘 Section 5: BI Dashboard Integration (Power BI/Tableau)

• Import data from Excel or CSV
• Use calculated fields to show mean, median, or percent change
• Visualize trends with line charts, bar plots, filters
• Create interactive dashboards for decision makers

📘 Section 6: Case Study – A/B Testing a Website Feature

🧪 Problem:

Does showing a product video increase conversion rates?

• Group A: Sees only text
• Group B: Sees video + text

python

group_A = [12, 15, 14, 13, 12]  # conversions out of 100 visitors

group_B = [18, 20, 19, 21, 22]

from scipy.stats import ttest_ind

t_stat, p_val = ttest_ind(group_B, group_A)

print("p-value:", p_val)

If p_val < 0.05: Evidence that video improved conversions.

📘 Section 7: Statistical Workflow in Python

✅ A typical statistical analysis pipeline:

python

# 1. Import and clean data

df = pd.read_csv('data.csv')

df.dropna(inplace=True)

# 2. Descriptive overview

print(df.describe())

sns.histplot(df['Revenue'])

# 3. Hypothesis test

from scipy.stats import ttest_1samp

t_stat, p_val = ttest_1samp(df['Revenue'], 10000)

# 4. Result interpretation

if p_val < 0.05:

    print("Revenue significantly differs from $10,000")

📘 Section 8: Best Practices

📋 Summary Table: Tools & Tasks