Skip to main content

Guided Case Studies

Skill Level: Beginner

In this page, you will complete two guided projects:

  1. Student Performance (exam scores)
  2. Better Life Index (country-level quality of life and life satisfaction)

Both will walk you through loading data, simple cleaning, basic analysis, visualization, and interpretation.

Guided Case Study 1: Student Performance​

Project Overview​

In this guided project, you will act as a data analyst helping a school understand which factors relate to student exam performance.

Main question:
How do factors like gender, lunch type, and test preparation relate to student scores?

You will:

  • load and explore a real dataset,
  • compute summary statistics,
  • create visualizations,
  • and write short interpretations of what you find.

Dataset and Setup​

We will use the Students Performance in Exams dataset from Kaggle:

Step 1: Download the Dataset​

  1. Open the Kaggle dataset page.
  2. Download StudentsPerformance.csv to your computer.

Step 2: Upload to Google Colab​

  1. Open a new notebook in Google Colab.
  2. Use the Upload button in the left file browser to upload StudentsPerformance.csv.

Step 3: Load with Pandas​

import pandas as pd

df = pd.read_csv("StudentsPerformance.csv")
df.head()

Check that the first few rows look reasonable.

Understand the Columns​

This dataset includes columns like:

  • gender
  • race/ethnicity
  • parental level of education
  • lunch
  • test preparation course
  • math score, reading score, writing score (column names may vary slightly)

Quick Exercise​

In your own words, write down 2–3 questions you might answer using these columns.
Examples:

  • “Do students who complete test preparation have higher scores?”
  • “Is there a difference in average scores by lunch type?”

Simple Cleaning​

Even if the dataset is mostly clean, it’s useful to:

Step 1: Check Missing Values​

df.isna().sum()

If all zeros, you can move on. If not, note which columns have missing data.

Step 2: Rename Columns (Optional)​

To make column names easier to work with:

df = df.rename(columns={
    "race/ethnicity": "race_ethnicity",
    "parental level of education": "parent_education",
    "test preparation course": "test_preparation"
})

Step 3: Create an Average Score Column​

df["avg_score"] = df[["math score", "reading score", "writing score"]].mean(axis=1)
df.head()

This gives you a single summary score for each student.

Descriptive Statistics​

Overall Summary​

df.describe()

Look at the summary for math score, reading score, writing score, and avg_score.

Grouped Averages​

By gender:

df.groupby("gender")["avg_score"].mean()

By lunch type:

df.groupby("lunch")["avg_score"].mean()
Interpretation Prompts​
  • Which gender group has the higher average avg_score?
  • How do average scores differ between lunch types?

Write 1–2 sentences for each question based on the numbers you see.

Core Visualizations​

We will create three key charts and interpret them.

1. Bar Chart: Average Score by Gender​

import matplotlib.pyplot as plt

df.groupby("gender")["avg_score"].mean().plot(kind="bar")
plt.title("Average Score by Gender")
plt.ylabel("Average Score")
plt.xlabel("Gender")
plt.show()

Your task:
Describe what this chart suggests in 2–3 sentences.
Consider:

  • Which bar is higher?
  • Is the difference large or small?

2. Boxplot: Score Distribution by Lunch Type​

import seaborn as sns
import matplotlib.pyplot as plt

sns.boxplot(data=df, x="lunch", y="avg_score")
plt.title("Score Distribution by Lunch Type")
plt.xlabel("Lunch Type")
plt.ylabel("Average Score")
plt.show()

Your task:

  • Which lunch group seems to have higher scores overall?
  • Does one group show a wider spread in scores?

Write a short paragraph explaining what you see.

3. Scatter Plot: Math vs Reading Scores​

sns.scatterplot(data=df, x="math score", y="reading score")
plt.title("Math Score vs Reading Score")
plt.xlabel("Math Score")
plt.ylabel("Reading Score")
plt.show()

Your task:

  • Does there appear to be a positive relationship between math and reading scores?
  • Do higher math scores tend to come with higher reading scores?

Explain briefly using the pattern of points.

Guided Questions​

Use your tables and charts to answer these:

  1. How do average scores differ by gender?
  2. How do average scores differ by lunch type?
  3. Does completing the test preparation course appear to be associated with higher scores? 
    df.groupby("test_preparation")["avg_score"].mean()
  4. Do math and reading scores show evidence of a positive relationship?

Write 2–3 sentences for each question.

Wrap-Up and Reflection​

Consider these reflection prompts:

  • If the school had limited resources, which group would you prioritize for extra support, based on your analysis?
  • What limitations does this dataset have (for example, missing context, only one school, etc.)?
  • What additional data would you want to collect to make stronger conclusions?

Write a short reflection (one paragraph) summarizing what you learned about student performance in this case study.

Guided Case Study 2: Better Life Index (Quality of Life)​

Project Overview​

In this guided project, you will analyze country-level quality of life and life satisfaction indicators to understand how different dimensions (such as income, health, and work-life balance) relate to overall wellbeing.

Main question:
How do different aspects of life (income, health, community, etc.) relate to life satisfaction across countries?

You will:

  • load and clean a real-world dataset,
  • handle column names and missing values,
  • compute summary statistics,
  • create visualizations,
  • and interpret patterns across countries.

Dataset and Setup​

We will use the Better Life Index 2024 dataset from Kaggle:

Download the main CSV file (for example, better_life_index_2024.csv, depending on the exact file name on Kaggle).

Step 1: Download the Dataset​

  1. Open the Kaggle dataset page.
  2. Download the Better Life Index CSV file to your computer.

Step 2: Upload to Google Colab​

  1. Open a new notebook in Google Colab.
  2. Upload your Better Life Index CSV file.

Step 3: Load with Pandas​

import pandas as pd

bli = pd.read_csv("better_life_index_2024.csv")  # adjust filename as needed
bli.head()

Check the first few rows and note the columns.

Inspect and Clean the Data​

This dataset typically includes:

  • a country column,
  • an overall life satisfaction or better life index score,
  • multiple indicator columns (income, jobs, health, education, environment, etc.).

Step 1: Check Column Names​

bli.columns

Identify key columns, for example:

  • Country
  • Life Satisfaction
  • Income
  • Jobs
  • Health
  • Education
  • Environment
  • Work-Life Balance
  • Community

(The exact names may differ; adjust based on what you see.)

Step 2: Rename Columns for Easier Use​

Create cleaner, Python-friendly names:

bli = bli.rename(columns={
    "Country": "country",
    "Life Satisfaction": "life_satisfaction",
    "Income": "income",
    "Jobs": "jobs",
    "Health": "health",
    "Education": "education",
    "Environment": "environment",
    "Work-Life Balance": "work_life_balance",
    "Community": "community"
    # add or adjust keys based on the actual dataset
})

bli.head()

Step 3: Check for Missing Values​

bli.isna().sum()

Note which columns have missing data.

For this guided project, we’ll create a cleaned version that drops rows missing key indicators:

key_cols = ["life_satisfaction", "income", "health", "education", "community"]
bli_clean = bli.dropna(subset=key_cols)

bli_clean.isna().sum()
bli_clean.shape

You now have a cleaned dataset called bli_clean.

Basic Descriptive Statistics​

Overall Summary​

bli_clean.describe()

Look at:

  • average life satisfaction,
  • average income,
  • average health and education scores.

Top and Bottom Countries by Life Satisfaction​

Top 10:

top10_bli = bli_clean.sort_values("life_satisfaction", ascending=False).head(10)
top10_bli[["country", "life_satisfaction"]]

Bottom 10:

bottom10_bli = bli_clean.sort_values("life_satisfaction").head(10)
bottom10_bli[["country", "life_satisfaction"]]
Interpretation Prompts​
  • Which countries appear at the top and bottom of the life satisfaction list?
  • Do you recognize any patterns (for example, similar regions or economic levels)?

Write a short comment on what you notice.

Core Visualizations​

We will create a few charts to understand relationships between life satisfaction and key indicators.

1. Bar Chart: Top 10 Countries by Life Satisfaction​

import matplotlib.pyplot as plt

plt.figure(figsize=(8, 5))
plt.barh(top10_bli["country"], top10_bli["life_satisfaction"])
plt.gca().invert_yaxis()
plt.title("Top 10 Countries by Life Satisfaction")
plt.xlabel("Life Satisfaction")
plt.show()

Your task:

  • Which countries stand out at the top?
  • How different are their scores from one another?

2. Scatter Plot: Income vs Life Satisfaction​

import seaborn as sns
import matplotlib.pyplot as plt

sns.scatterplot(
    data=bli_clean,
    x="income",
    y="life_satisfaction"
)
plt.title("Income vs Life Satisfaction")
plt.xlabel("Income")
plt.ylabel("Life Satisfaction")
plt.show()

Your task:

  • Does life satisfaction tend to increase with income?
  • Are there countries that have relatively high satisfaction but moderate income, or the opposite?

Describe the general pattern.

3. Scatter Plot: Health vs Life Satisfaction​

sns.scatterplot(
    data=bli_clean,
    x="health",
    y="life_satisfaction"
)
plt.title("Health vs Life Satisfaction")
plt.xlabel("Health")
plt.ylabel("Life Satisfaction")
plt.show()

Your task:

  • Does better health seem associated with higher life satisfaction?
  • How strong does the relationship appear?

Write a brief interpretation.

4. Scatter Plot: Community vs Life Satisfaction​

sns.scatterplot(
    data=bli_clean,
    x="community",
    y="life_satisfaction"
)
plt.title("Community vs Life Satisfaction")
plt.xlabel("Community")
plt.ylabel("Life Satisfaction")
plt.show()

Your task:

  • Do stronger community scores appear to relate to higher life satisfaction?
  • Are there countries with strong community but moderate satisfaction?

Guided Questions​

Use your tables and charts to answer these:

  1. How would you describe the relationship between income and life satisfaction?
  2. How would you describe the relationship between health and life satisfaction?
  3. How does community relate to life satisfaction in this dataset?
  4. Which factor (income, health, community) appears most clearly related to life satisfaction in your charts?

Write 2–3 sentences for each question.

Data Cleaning Reflection​

Think about the cleaning steps you took:

  • renaming columns,
  • dropping rows with missing values in key columns.

Reflect briefly:

  • Why did renaming columns make your analysis easier?
  • What are the pros and cons of dropping rows versus trying to fill missing values?

Write a short paragraph summarizing your thoughts on data cleaning in this project.

Wrap-Up and Reflection​

Consider:

  • If you were advising policymakers on improving quality of life, which indicators would you emphasize based on your analysis?
  • What important aspects of wellbeing might be missing from this dataset?
  • What additional data (for example, inequality, mental health, environmental quality) would you want to make stronger conclusions?

Write one paragraph reflecting on what you learned about country-level quality of life and life satisfaction.

Next Steps​

After completing both guided case studies:

  • Choose one Beginner Project from the Practice Projects page.
  • Apply the same workflow: load data, clean/rename columns, explore with summaries and charts, and write interpretations.
  • Publish your notebooks and results on GitHub as part of your growing portfolio.

Question for you: for the Better Life Index guided project, do you want to add one optional “challenge” where students compare life satisfaction across a few chosen regions or income groups to push their grouping/aggregation skills a bit further?