Sample size
When full information is hidden or inaccessible, recognize the logical relationship between a sufficient number of chances and a sufficiently large sample to reasonably represent something.
K–2 Competencies
Classroom resources
Data Science Starter Kit Module 4: Drawing Conclusions - Interpreting Problems and Results
Welcome to one of the most critical skills in data science—learning how to draw valid conclusions from data and understanding what those conclusions can and cannot tell us. This module focuses on the thinking skills that separate good data science from misleading claims.🔗
Interpreting Problems and Results isn’t about complex statistical tests or advanced mathematics. It’s about developing the intellectual honesty and critical thinking skills to say, “Based on this data, here’s what we can reasonably conclude, here’s what we’re not sure about, and here’s what we still need to investigate.”
3–5 Competencies
Recognize that in a scenario of random chance (e.g., dice rolls, jar of jelly beans), too few trials can skew conclusions. e.g., flipping a coin twice and getting heads both times doesn't mean it's always heads and more flips will provide a clearer picture
Classroom resources
Data Science Starter Kit Module 4: Drawing Conclusions - Interpreting Problems and Results
Welcome to one of the most critical skills in data science—learning how to draw valid conclusions from data and understanding what those conclusions can and cannot tell us. This module focuses on the thinking skills that separate good data science from misleading claims.🔗
Interpreting Problems and Results isn’t about complex statistical tests or advanced mathematics. It’s about developing the intellectual honesty and critical thinking skills to say, “Based on this data, here’s what we can reasonably conclude, here’s what we’re not sure about, and here’s what we still need to investigate.”
6–8 Competencies
Recognize that a sample must be sufficiently large to well-represent a broader population, based on the concept of the Law of Large Numbers. e.g., flipping a coin 10 times might give 7 heads, but 1000 flips will trend towards 50/50
Identify examples of too-small sample sizes in the media or other real-world examples. e.g., medical drug drials, prior debunked research
Classroom resources
Data Science Starter Kit Module 4: Drawing Conclusions - Interpreting Problems and Results
Welcome to one of the most critical skills in data science—learning how to draw valid conclusions from data and understanding what those conclusions can and cannot tell us. This module focuses on the thinking skills that separate good data science from misleading claims.🔗
Interpreting Problems and Results isn’t about complex statistical tests or advanced mathematics. It’s about developing the intellectual honesty and critical thinking skills to say, “Based on this data, here’s what we can reasonably conclude, here’s what we’re not sure about, and here’s what we still need to investigate.”
9–10 Competencies
Recognize there are formal methods to determine the minimum sample size needed to make a well-supported claim about a population.
Explain “statistical power” of a statistical test as the general probability that an outcome “lands” more “extremely,” beyond an arbitrary pivotal value set for statistical significance that a researcher chooses.
Explain “statistical power” as the probability that a statistical test properly detects a real effect when one exists.
Classroom resources
Data Science Starter Kit Module 4: Drawing Conclusions - Interpreting Problems and Results
Welcome to one of the most critical skills in data science—learning how to draw valid conclusions from data and understanding what those conclusions can and cannot tell us. This module focuses on the thinking skills that separate good data science from misleading claims.🔗
Interpreting Problems and Results isn’t about complex statistical tests or advanced mathematics. It’s about developing the intellectual honesty and critical thinking skills to say, “Based on this data, here’s what we can reasonably conclude, here’s what we’re not sure about, and here’s what we still need to investigate.”
11–12 Competencies
Make an informal power analysis for an analysis or experimental setup using real-world data and a hypothesis, including claims about the 1) Effect Size 2) Sample Size 3) Statistical Significance and 4) Statistical Power.
Use the simple equation Power = 1 - β to visually show the difference between a normal distribution of outcomes and an abnormal distribution of outcomes.
Classroom resources
Data Science Starter Kit Module 4: Drawing Conclusions - Interpreting Problems and Results
Welcome to one of the most critical skills in data science—learning how to draw valid conclusions from data and understanding what those conclusions can and cannot tell us. This module focuses on the thinking skills that separate good data science from misleading claims.🔗
Interpreting Problems and Results isn’t about complex statistical tests or advanced mathematics. It’s about developing the intellectual honesty and critical thinking skills to say, “Based on this data, here’s what we can reasonably conclude, here’s what we’re not sure about, and here’s what we still need to investigate.”
Advanced Competencies
Make a formal Power Analysis by identifying a sufficient sample size for a real-world data exploration. Students should mathematically isolate “n” in a t-test or z-test, and estimate Power with a software tool.
Classroom resources
Support other teachers by sharing a resource
Do you have a lesson plan, video, or tip that could help others teaching this topic?
Developed by our coalition
.png){kind=logo}
Coalition organizers
Share feedback on the Learning Progressions
Your feedback helps us improve these progressions for teachers around the world. Thank you!
Share feedback on the Learning Progressions
Your feedback helps us improve these progressions for teachers around the world. Thank you!
Share a classroom resource
Suggesting a resource helps students around the world learn essential data science skills.