Case Study 2: The Graduation Speech — Designing a Statistical Literacy Campaign

The Setup

You've been asked to give a three-minute speech at your university's data science club end-of-year banquet. The topic: "What I Wish Everyone Knew About Statistics."

Your audience is a mix of data science students, business majors, a few liberal arts students who wandered in for the free food, and several professors from non-quantitative departments. Some of them have never taken a statistics course. Others have taken three.

Your goal isn't to teach them statistics. It's to convince them that statistical thinking matters — and to give them three concrete tools they can start using immediately, no formulas required.

This case study asks you to design that speech, drawing on everything you've learned.

Part 1: The Hook

Every good speech starts with a story that captures attention.

Your Task

Choose ONE of the following opening scenarios and write the first 60 seconds of your speech (approximately 150-180 words). Your opening should make the audience lean forward.

Option A: The Medical Scare

A headline reads: "New Study Finds That Drinking Coffee Doubles Your Risk of Heart Disease." You see it shared on social media 50,000 times. Your mom texts you about it. Here's what the headline didn't tell you...

Option B: The AI Verdict

A judge uses an algorithm to decide whether a 22-year-old defendant gets bail or sits in jail for six months awaiting trial. The algorithm says "high risk." The defendant's lawyer asks: "How accurate is this tool?" The answer depends on a question that most people in this room have never thought to ask...

Option C: The Election Night Surprise

A polling firm predicts that Candidate A will win with 52% of the vote, plus or minus 3 percentage points. Candidate B wins with 51%. The next morning, headlines scream: "Pollsters Got It Wrong Again!" But did they?

Option D: Your Own Story

Use a real experience from your own life where statistical thinking changed your understanding of something important.

Analysis Questions

(a) Write your 150-180 word opening for the scenario you chose.

(b) Identify which of the six recurring themes your opening most directly addresses. Explain your choice.

(c) Which chapters from the textbook are most relevant to your chosen scenario? List at least three, and for each, name the specific concept that applies.

Part 2: The Three Tools

After your hook, you want to give your audience three simple tools they can use to think statistically — without any formulas.

Here are the six themes of this textbook, reframed as questions anyone can ask:

Your Task

Choose three of these questions as your "Three Tools for Statistical Thinking." For each one:

(a) Write a 100-word explanation of the tool, suitable for a non-technical audience. No jargon allowed.

(b) Give a concrete, real-world example where using this tool would prevent a bad decision or reveal a hidden truth.

(c) Connect the tool to at least two specific chapters from this textbook.

Part 3: The Real-World Challenge

Now let's apply your three tools to a complex, realistic scenario.

Scenario: The School District Data

The Lakewood School District has released data showing that students who participated in an after-school tutoring program scored, on average, 15 points higher on the state math exam than students who didn't participate. The district is using this data to justify a $2 million expansion of the program.

Here's what the data shows:

The two-sample t-test gives t = 15.4, p < 0.001, Cohen's d = 0.92.

A local newspaper headline reads: "After-School Tutoring Boosts Math Scores by 15 Points, Study Finds"

Analysis Questions

(a) Apply each of your three tools to this scenario. What questions would each tool prompt you to ask?

(b) The result is statistically significant (p < 0.001) with a large effect size (d = 0.92). Does this mean the tutoring program caused the improvement? Use concepts from Chapters 4 and 22 to explain your reasoning.

(c) List at least four confounders that could explain the difference between the tutored and non-tutored groups. For each, explain the direction of bias (would controlling for it increase or decrease the estimated effect?).

(d) The school district plans to expand the program based on this data. What study design would you recommend to establish a causal effect? Be specific about randomization, sample size, and potential ethical concerns.

(e) A school board member says: "The confidence interval for the tutored group is (77.0, 79.6) and for the non-tutored group is (62.0, 64.2). These intervals don't overlap, so the tutoring works." Is this reasoning correct? Why or why not? (Hint: recall the discussion of overlapping CIs from Chapter 16.)

(f) Write a 200-word response to the newspaper headline, suitable for a letter to the editor, that applies statistical thinking without being dismissive of the program.

Part 4: The Closing

Every great speech needs a memorable closing. Here are the closing lines from several chapters of this textbook:

"Statistics is the most useful subject you'll study in college." (Chapter 1)

"You've come an extraordinary distance. One chapter remains." (Chapter 27)

"Statistics does not require certainty — only curiosity, honesty, and the courage to let data change your mind." (Chapter 28)

Your Task

(a) Write the closing 100 words of your speech. It should be personal, specific, and leave your audience thinking about statistics differently than when they sat down.

(b) Why did you choose these particular words? What idea or feeling do you want lingering in the room after you sit down?

Part 5: The Meta-Question

(a) This case study asked you to explain statistics to non-statisticians. What was the hardest part? What concepts were easiest to explain in plain language, and which resisted simplification?

(b) Chapter 25 argued that "communication is the superpower" — that analysis without clear communication is analysis that doesn't exist. After completing this case study, do you agree? Why or why not?

(c) Imagine you're revisiting this case study five years from now. What do you think will have changed about the role of statistics in everyday life? What do you think will be the same?

(d) One of the themes of this textbook is that statistical thinking is not just a technical skill but a civic responsibility. In 100-150 words, explain what this means to you personally. Be specific — don't just agree with the statement; explain why you agree (or disagree) and give an example from your own experience.

Assessment Rubric

If this case study is being graded, here's what excellent work looks like: