๐Ÿ“ Student Worksheet โ€” Session 4: When Data Deceives

Data Science for Young Minds ยท Grade 5 ยท Ages 10โ€“11
๐Ÿ“š Part 1 โ€” Vocabulary
Cherry-Picking
Survivorship Bias
Framing Effect
Simpson's Paradox
Confirmation Bias
Subgroup
๐Ÿ” Part 2 โ€” Name That Trick

Identify the deception technique in each scenario. Write: Cherry-Picking / Survivorship Bias / Framing Effect / Simpson's Paradox

ScenarioTechnique
"Our new exercise program works! We have 50 success stories to prove it." (The 200 people who quit early are not mentioned.)
"Our toothpaste is 97% cavity-free!" vs. "3% of users still developed cavities."
"Our school improved! Overall test scores went from 55% to 60%." But scores for both high-achieving AND low-achieving students both went down by 5%.
A company shows a bar graph of quarterly profits โ€” only showing Q1 and Q4 (their two best quarters).
"All the famous musicians practiced 10,000 hours. So practicing 10,000 hours will make you famous!"
๐Ÿ”„ Part 3 โ€” Simpson's Paradox Analysis

Study this data about two reading programs at Lincoln School:

Student GroupProgram X pass rateProgram Y pass rate
Grade 4 students80% (40 out of 50)75% (15 out of 20)
Grade 5 students60% (12 out of 20)50% (25 out of 50)
Overall combined74% (52/70)57% (40/70)

Answer these questions:

1. For Grade 4 students, which program is better? _______________

2. For Grade 5 students, which program is better? _______________

3. Looking at the overall combined rate, which program appears better? _______________

4. Is this an example of Simpson's Paradox? Circle: YES / NO

5. Explain in 2โ€“3 sentences why the combined number can be misleading here:

6. If you were advising the school principal, which program would you recommend โ€” and why?

๐Ÿ•ต๏ธ Part 4 โ€” Data Detective Case Studies

For each case: identify the technique, find the flaw, and write a "detective question" that exposes it.

Case 1: "Our city's crime rate dropped 40% this year!"

The city changed how they count crime this year, excluding minor incidents that were previously included. The absolute number of serious crimes stayed the same.

Technique used: _______________________

The flaw:

My detective question:

Honest restatement:

Case 2: "Students who do homework every night get A grades โ€” so homework causes good grades!"

The data comes from a survey of top-performing students. Students who struggled and did homework but still got poor grades were not included in the analysis.

Technique used: _______________________

The flaw:

My detective question:

Honest restatement:

Case 3: "Warning: this medication has a 1-in-20 chance of causing liver damage."

A competing pharmaceutical company advertises their drug instead: "Our medication is safe for 95% of patients!" Both statistics refer to the same risk level (5%).

Technique used: _______________________

Explain why the framing creates different impressions even though the data is identical:

My detective question:

Case 4 โ€” Challenge: Baseball batting averages

Player A has a higher batting average than Player B against both left-handed pitchers AND right-handed pitchers. But Player B has a higher overall batting average for the season. Explain how this is possible and what it means for who is actually the better batter.

Technique at work: _______________________

Explanation:

Who is actually the better batter, and why?

โœ๏ธ Part 5 โ€” Reflection

Which deception technique was hardest to spot โ€” and why? What question would you ask to catch it in the future?

Think of one real-world situation where these tricks could cause serious harm (not just confusion). Describe it.

๐Ÿ  Take-Home Challenge โ€” Deception Spotter

Find one real advertisement, news headline, or statistic that uses one of today's deception techniques.

โ˜ Source: ___________________________________________
โ˜ Write out the claim or headline
โ˜ Name the deception technique being used
โ˜ Write the detective question that exposes it
โ˜ Write a more honest version of the claim

My honest restatement of the claim I found: