Drawing Inferences from Samples

Active learning helps students grasp the abstract shift from describing samples to making population inferences. When they manipulate real data, they confront variability and bias directly, which builds durable understanding of why inferences are estimates, not certainties. Collaborative structures let students hear peers articulate uncertainty in ways that teacher explanation alone cannot match.

• 7th Grade
• Mathematics
• 3 activities
• 20 min–35 min

Learning Objectives

1. Calculate the proportion of a characteristic within a random sample to estimate its proportion in a population.
2. Compare the results of inferences drawn from random samples versus non-random samples to justify the reliability of a sampling method.
3. Construct a plausible inference about a population's characteristic based on given sample data, using appropriate language for estimation.
4. Analyze the relationship between sample size and the confidence one can have in an inference about a population.
5. Evaluate the potential bias in an inference based on the description of the sampling method used.

Think-Pair-Share

⏱ 20 min·Pairs

Think-Pair-Share: Making the Inference

Provide three sample data sets with varying collection methods. Students individually write one inference from each and rate how confident they are in it. Partners compare ratings and discuss what drives their confidence levels before a whole-class share.

Prepare & details

Analyze how to make predictions about a population based on sample data.

Facilitation Tip: During Think-Pair-Share: Making the Inference, circulate and listen for students using absolute language like 'the population definitely…' and prompt them to revise with 'likely' or 'estimated'.

Setup: Standard classroom seating; students turn to a neighbor

Materials: Discussion prompt (projected or printed), Optional: recording sheet for pairs

• Understand
• Apply
• Analyze
• Self-Awareness
• Relationship Skills

Jigsaw

⏱ 35 min·Small Groups

Jigsaw: Sample Quality Review

Divide the class into four groups, each analyzing a different fictional survey with a described sampling method. Each group becomes 'experts' on their survey's strengths and weaknesses, then regroups in mixed teams to compare all four surveys and rank them by inference reliability.

Prepare & details

Justify the reliability of an inference based on the sampling method used.

Facilitation Tip: During Jigsaw: Sample Quality Review, assign each expert group a different sampling flaw so they must articulate why the flaw undermines inference quality.

Setup: Flexible seating for regrouping

Materials: Expert group reading packets, Note-taking template, Summary graphic organizer

• Understand
• Analyze
• Evaluate
• Relationship Skills
• Self-Management

Inquiry Circle

⏱ 25 min·Pairs

Inference Card Match

Create two sets of cards: sample descriptions and corresponding inferences. Some inference cards are plausible; others overreach. Students match each sample to the most defensible inference and justify their choices in writing before discussing as a class.

Prepare & details

Construct a plausible inference about a population given a set of sample data.

Facilitation Tip: During Inference Card Match, require students to justify each match using both sample statistic and variability language before they can declare a pair correct.

Setup: Groups at tables with access to source materials

Materials: Source material collection, Inquiry cycle worksheet, Question generation protocol, Findings presentation template

• Analyze
• Evaluate
• Create
• Self-Management
• Self-Awareness

Teaching This Topic

Teachers should avoid presenting inference as a single correct answer and instead normalize the language of uncertainty. Use multiple small samples from the same population early on to make variability visible before formal definitions. Emphasize that random sampling does not eliminate bias; it only makes bias unpredictable, which is preferable to systematic bias.

What to Expect

By the end of the activities, students will qualify inferences with appropriate language, identify sampling limitations, and explain how random selection supports valid conclusions. They will also recognize that multiple samples from the same population produce different results, reinforcing the idea of sampling variability.

Watch Out for These Misconceptions

Common Misconception

During Think-Pair-Share: Making the Inference, watch for students stating absolute conclusions from sample data.

What to Teach Instead

Redirect by asking them to revise their statements using probability language: 'Can you change “the population prefers” to “the sample suggests that it is likely that the population prefers”?' Have them practice this revision in their pairs before sharing with the class.

Common Misconception

During Inference Card Match, watch for students pairing sample proportions directly to population proportions without accounting for variability.

What to Teach Instead

Require students to reference the sample size and margin of error when matching cards, or provide a range card (e.g., 20%-30%) instead of a single percentage to emphasize that exact matches are unlikely.

Assessment Ideas

Quick Check

After Think-Pair-Share: Making the Inference, collect each pair’s written inference and justification. Look for use of probability language and explicit mention of sample-to-population reasoning.

Discussion Prompt

During Jigsaw: Sample Quality Review, have each expert group present their sampling flaw and ask the class to vote on which flaw poses the greatest threat to inference validity. Circulate and listen for students identifying direction of bias (over- or under-representation).

Exit Ticket

After Inference Card Match, ask students to write one sentence explaining why two samples from the same population might yield different results and one sentence describing how a larger sample could improve their inference.

Extensions & Scaffolding

• Challenge: Provide a biased sampling scenario and ask students to design a better method, then simulate it with a random sample generator.
• Scaffolding: Give students sentence frames for qualifying inferences: 'Based on this random sample, I estimate that ___ of the population ____. This is not certain because ____.'
• Deeper: Have students collect their own small samples from the school population, calculate statistics, and compare across different sample sizes to observe how variability decreases as sample size increases.

Key Vocabulary

Inference	A conclusion reached on the basis of evidence and reasoning, specifically about a population based on sample data.
Population	The entire group of individuals or objects that a study is interested in, about which conclusions are to be drawn.
Sample	A subset of individuals or objects selected from a population, used to make inferences about the entire population.
Random Sample	A sample where every member of the population has an equal chance of being selected, which helps ensure the sample is representative.
Characteristic	A specific feature or attribute of individuals or objects within a population or sample, such as preference for a certain food or color.