Surveys and Observational Studies
A free Statistics and Data Analysis lesson from the “Collecting Data” unit, with a worked example and practice problems including step-by-step solutions.
Surveys collect responses, and observational studies record what happens naturally. They can reveal patterns and associations, but without random assignment they usually cannot prove causation. This lesson builds the habit of reading the context first, choosing the right statistical tool, calculating carefully, and then writing what the result means. By the end, students should be able to do the computation and explain why that computation answers the question.
What you'll learn
- Identify survey wording issues
- Describe observational study limits
- Explain association without overclaiming causation
Worked example
Problem. A study records students' sleep time and quiz scores without assigning sleep schedules. What type of study is this?
- Worked Example: First identify exactly what the question is asking: A study records students' sleep time and quiz scores without assigning sleep schedules. What type of study is this?
- Compare each answer choice with the calculation or rule, and eliminate choices that do not satisfy the condition.
- The researcher only records what happens naturally.
- No treatment is assigned, so this is observational.
Answer: observational study
Practice problems
1. Practice case A: A question says, 'Would you support the helpful new schedule?' What is the concern?
Choices: the sample is automatically random · the study is double-blind · the result proves causation · the wording is biased
Show solution
- Warm-up: First identify exactly what the question is asking: A question says, 'Would you support the helpful new schedule?' What is the concern?
- Compare each answer choice with the calculation or rule, and eliminate choices that do not satisfy the condition.
- Loaded words push the respondent toward an answer.
- Survey questions should be neutral.
- Verify the selected choice by checking that it satisfies the original prompt and that the other choices fail the same test.
Answer: the wording is biased
2. Practice case B: Which wording is best for measuring opinion without leading?
Choices: How satisfied are you with the current lunch schedule? · Do you love the improved lunch schedule? · Why is the lunch schedule so inconvenient? · Should we keep the obviously better schedule?
Show solution
- Warm-up: First identify exactly what the question is asking: Which wording is best for measuring opinion without leading?
- Compare each answer choice with the calculation or rule, and eliminate choices that do not satisfy the condition.
- Neutral wording does not suggest the desired answer.
- It lets respondents answer honestly.
- Verify the selected choice by checking that it satisfies the original prompt and that the other choices fail the same test.
Answer: How satisfied are you with the current lunch schedule?
3. Practice case C: A researcher compares students who already participate in sports with those who do not. No one is assigned to a group. What can this study usually show?
Choices: that all variables are categorical · association, not definite causation · cause and effect with certainty · the exact population parameter
Show solution
- Warm-up: First identify exactly what the question is asking: A researcher compares students who already participate in sports with those who do not. No one is assigned to a group. What can this study usually show?
- Compare each answer choice with the calculation or rule, and eliminate choices that do not satisfy the condition.
- Observational studies can reveal patterns.
- Without random assignment, causation is harder to justify.
- Verify the selected choice by checking that it satisfies the original prompt and that the other choices fail the same test.
Answer: association, not definite causation
4. Practice case D: A teacher emails a survey to a full class, but only a few students answer. What is the main concern?
Choices: random assignment · a larger margin of error only · nonresponse bias · placebo effect
Show solution
- Warm-up: First identify exactly what the question is asking: A teacher emails a survey to a full class, but only a few students answer. What is the main concern?
- Compare each answer choice with the calculation or rule, and eliminate choices that do not satisfy the condition.
- People who do not respond may differ from those who do.
- That can bias the result.
- Verify the selected choice by checking that it satisfies the original prompt and that the other choices fail the same test.
Answer: nonresponse bias
5. Practice case E: A survey asks students to estimate screen time from memory. Which issue could affect accuracy?
Choices: random assignment · chi-square association · placebo control · recall error
Show solution
- Warm-up: First identify exactly what the question is asking: A survey asks students to estimate screen time from memory. Which issue could affect accuracy?
- Compare each answer choice with the calculation or rule, and eliminate choices that do not satisfy the condition.
- People may not remember details perfectly.
- That can affect survey data.
- Verify the selected choice by checking that it satisfies the original prompt and that the other choices fail the same test.
Answer: recall error
Practice this interactively with instant feedback and an AI tutor.
Practice Surveys and Observational Studies Take the free placement check