10 Assignment 9: Streaming Analytics — Understanding Platform Popularity Across Age Groups

Chi-Square • Residual Analysis • Contributions • Cramer’s V

11 Overview

The Streaming Analytics Division (SAD) has hired you as their new Data Analyst. Your mission is to determine whether age group influences streaming platform preference.

The company wants to know whether certain platforms — Netflix, Hulu, Disney+, Amazon, or Other — appeal more strongly to specific age demographics (18–25, 26–40, 41+). Your analysis will guide targeted marketing strategies, promotional campaigns, and content investment decisions.

Using simulated survey data, you will conduct a fully reproducible Quarto analysis to determine whether Age Category and Platform Preference are statistically related.

You will:

Construct and interpret contingency tables
Perform a Chi-Square test of independence
Examine observed, expected, and residual values
Decompose the Chi-Square statistic into contributions
Visualize contributions using a heatmap
Compute and interpret Cramer’s V
Communicate findings clearly in plain language

12 Learning Objectives

By completing this assignment, you will be able to:

Clean and summarize categorical data
Construct contingency tables
Conduct and interpret a Chi-Square test of independence
Interpret observed, expected, and residual frequencies
Decompose the Chi-Square statistic into cell-level contributions
Visualize categorical data using stacked and clustered bar charts
Calculate and interpret Cramer’s V
Communicate statistical findings in accessible language

13 Textbook Connection

This assignment builds directly from Chapter 6: Analyzing Categorical Data in Reproducible Research Using R.

Students are encouraged to review the chapter before beginning this assignment, as it provides the conceptual foundation and reproducible workflow demonstrated here.

14 Submission Instructions

Submit two files:

Your .qmd file
Your knitted and published .html file

Your document must:

Knit without errors
Suppress warnings and messages in the final HTML
Include clearly labeled sections
Include all required tables and visualizations
Include full statistical interpretations
Use at least one inline R expression

15 Assignment Tasks

15.1 1. Data Preparation

You have been provided with survey data including:

Age Category (18–25, 26–40, 41+)
Preferred Streaming Platform (Netflix, Hulu, Disney+, Amazon, Other)

15.1.1 Required Tasks

You must:

Inspect the dataset
Compute total counts for Age Category
Compute total counts for Platform Preference
Construct a contingency table showing the joint distribution of Age Category and Platform Preference

Briefly describe what the contingency table represents and why it is useful for categorical analysis.

15.2 2. Visualization

Create two visualizations using ggplot2.

15.2.1 Required Visualization 1

A stacked bar chart showing the proportions of platform preferences within each age group.

15.2.2 Required Visualization 2

A clustered (side-by-side) bar chart showing counts for each platform across age groups.

Both visualizations must:

Include descriptive titles
Include labeled axes
Include a legend
Use a professional or themed style (e.g., from ggthemes)

After each plot, provide 3–4 sentences interpreting what you observe.

15.3 3. Chi-Square Test of Independence

Conduct a Chi-Square test of independence to determine whether Age Category and Platform Preference are related.

In your write-up, report and interpret:

The Chi-Square statistic (χ²)
Degrees of freedom (df)
The p-value

Then answer:

Is there a statistically significant relationship?
What does this result mean in plain language?

15.4 4. Observed, Expected, and Residual Values

From your Chi-Square test, extract:

Observed counts
Expected counts
Residuals

Discuss:

Which age–platform combinations show more people than expected
Which show fewer people than expected
Which residuals appear largest in magnitude

Explain what residuals tell us about patterns in the data.

15.5 5. Contributions to the Chi-Square Statistic

Break down the Chi-Square statistic into individual cell contributions.

Then:

Calculate the percentage contribution of each cell to the total χ² value
Identify which age–platform combinations contribute most to the overall result

Visualize these percentage contributions using a heatmap.

In your interpretation, explain:

Which cells drive the overall Chi-Square result
What this suggests about viewing habits across age groups

15.6 6. Effect Size (Cramer’s V)

Compute Cramer’s V to measure the strength of association between Age Category and Platform Preference.

Interpret the value using these guidelines:

0.00–0.10 → Very Weak
0.10–0.30 → Weak
0.30–0.50 → Moderate
Above 0.50 → Strong

Provide a short, plain-language explanation of the effect size.

15.7 7. Final Interpretation

Write a professional summary addressing:

Whether the relationship between Age and Platform is statistically significant
Which age groups and platforms are driving the relationship
The strength of the association (Cramer’s V)
What these findings mean in a real-world marketing context

Your summary should resemble a professional analytics report, not a textbook answer.

16 Deliverables

Your final report must include:

The contingency table
Both visualizations (stacked and clustered)
Chi-Square results (χ², df, p-value)
Observed, expected, and residual tables
Contribution table and heatmap
Cramer’s V calculation and interpretation
Final written summary

17 Guidelines for Interpretation

When discussing your results, always include:

Statistical significance — Is the relationship real?
Direction of pattern — Which combinations are over- or under-represented?
Strength of association — Using Cramer’s V
Plain-language meaning — What does this imply about streaming habits?

Avoid jargon without explanation.

18 Reflection

In 4–6 sentences, answer:

Why is the Chi-Square test appropriate for categorical data?
What is the difference between statistical significance and effect size?
Why is it important to examine residuals and contributions rather than only the p-value?

19 Reproducibility Practice

This assignment emphasizes transparent categorical analysis.

Your document must:

Construct tables programmatically
Avoid manual calculations outside your script
Use clearly labeled code chunks
Avoid hard-coding statistical results in narrative
Use inline R expressions where appropriate
Render cleanly to HTML

The goal is that another analyst could reproduce your entire Chi-Square analysis and verify every conclusion.

20 Publishing Instructions

Render your .qmd file to HTML.
Confirm warnings and messages do not appear in the final document.
Submit both your .qmd file and your published .html report.