Experimental vs Quasi-Experimental Design: Differences and Examples
Separate randomized experiments from quasi-experimental comparisons by asking who assigns treatment, whether assignment is random, and what causal claim is justified.
Experimental Vs Quasi Experimental Design: direct answer
Experimental vs quasi experimental design is primarily a question about treatment assignment. Randomized experiments assign treatment by chance; quasi-experimental comparisons rely on nonrandom mechanisms and therefore need stronger assumptions about comparability and confounding.
This page keeps the lesson centered on experimental vs quasi experimental design. Practice is included only after the method, assumptions, interpretation, and common decision points are explained.
Quick reference: Experimental vs Quasi-Experimental Design: Differences and Examples
| Quasi-experiment | A treatment comparison without random assignment. |
|---|
Concept mastery: Experimental vs Quasi-Experimental Design: Differences and Examples
Random assignment is the central dividing line
A randomized experiment deliberately assigns treatments by chance. A quasi-experimental design studies an intervention or exposure without full random assignment, often because policy, geography, timing, eligibility rules, or practical constraints determine who receives treatment. Both may compare treated and untreated groups, but their protection against confounding differs.
Natural experiments can create useful comparisons without researcher randomization
Sometimes an external event or rule creates groups that are plausibly comparable, such as a policy threshold or sudden implementation date. These situations can strengthen causal reasoning compared with a simple observational comparison, but the justification depends on the assignment mechanism and assumptions rather than on the label “natural experiment.”
Preexisting groups invite confounding
If students choose whether to enroll in a tutoring program, later score differences can reflect motivation, prior achievement, scheduling, family support, or other differences as well as tutoring. Statistical adjustment can address measured variables but cannot guarantee balance on unmeasured confounders the way random assignment is designed to do.
Before-and-after designs need a credible counterfactual
A change after an intervention does not automatically equal the intervention effect. Time trends, seasonality, maturation, outside events, or regression to the mean can also change outcomes. A comparison group following the same time period can improve the design when it represents what would likely have happened without treatment.
AP Statistics causal language should follow the design evidence
For current AP Statistics, the core distinction remains whether a study uses random assignment and therefore supports a causal treatment comparison. Quasi-experimental terminology can help understand real-world studies, but students should not use it to bypass the requirement to explain assignment, confounding, and scope.
A strong critique proposes a feasible design improvement
When randomization is impossible, improve the comparison by measuring important baseline variables, choosing a credible comparison group, examining pre-intervention trends, standardizing outcome measurement, and stating residual limitations. The purpose is not to claim equivalence with randomization but to make assumptions transparent.
12 worked experimental vs quasi experimental design cases
Worked case 1: Randomized vaccine trial
Scenario. Eligible participants are randomly assigned vaccine or comparator.
Reasoning. This is a randomized experiment; chance assignment is the key protection against systematic baseline differences.
Case 1: Randomized vaccine trial check: identify the exact statistical target in this scenario, verify the sign, denominator, assignment mechanism, event boundary, or model condition that controls the answer, and end with a conclusion whose scope matches the evidence. This check is specific to case 1; do not replace it with a memorized generic sentence.
Worked case 2: Policy by county
Scenario. One county adopts a policy while a neighboring county does not.
Reasoning. The comparison can be quasi-experimental, but county differences can confound the policy effect unless the design provides stronger justification.
Case 2: Policy by county check: identify the exact statistical target in this scenario, verify the sign, denominator, assignment mechanism, event boundary, or model condition that controls the answer, and end with a conclusion whose scope matches the evidence. This check is specific to case 2; do not replace it with a memorized generic sentence.
Worked case 3: Eligibility cutoff
Scenario. Aid is granted to applicants just below a score threshold but not just above it.
Reasoning. A threshold can create a quasi-experimental contrast when cases near the cutoff are otherwise comparable, but the assumption must be defended.
Case 3: Eligibility cutoff check: identify the exact statistical target in this scenario, verify the sign, denominator, assignment mechanism, event boundary, or model condition that controls the answer, and end with a conclusion whose scope matches the evidence. This check is specific to case 3; do not replace it with a memorized generic sentence.
Worked case 4: Self-selected program
Scenario. Employees decide whether to join a wellness program.
Reasoning. Self-selection makes a simple treated-versus-untreated comparison observational and vulnerable to motivation and health differences.
Case 4: Self-selected program check: identify the exact statistical target in this scenario, verify the sign, denominator, assignment mechanism, event boundary, or model condition that controls the answer, and end with a conclusion whose scope matches the evidence. This check is specific to case 4; do not replace it with a memorized generic sentence.
Worked case 5: Interrupted time series
Scenario. Monthly accident rates are recorded for years before and after a law change.
Reasoning. The time pattern can support a quasi-experimental argument, but concurrent events and trends remain alternative explanations.
Case 5: Interrupted time series check: identify the exact statistical target in this scenario, verify the sign, denominator, assignment mechanism, event boundary, or model condition that controls the answer, and end with a conclusion whose scope matches the evidence. This check is specific to case 5; do not replace it with a memorized generic sentence.
Worked case 6: Difference in differences
Scenario. Two regions have similar pre-policy trends; one adopts a policy and one does not.
Reasoning. Comparing changes can remove stable baseline differences, but credibility depends on the assumption that trends would have remained comparable without the policy.
Case 6: Difference in differences check: identify the exact statistical target in this scenario, verify the sign, denominator, assignment mechanism, event boundary, or model condition that controls the answer, and end with a conclusion whose scope matches the evidence. This check is specific to case 6; do not replace it with a memorized generic sentence.
Worked case 7: Natural lottery
Scenario. A real external lottery determines access to a program.
Reasoning. If the lottery genuinely controls access, the mechanism can approximate random assignment even though researchers did not run the lottery.
Case 7: Natural lottery check: identify the exact statistical target in this scenario, verify the sign, denominator, assignment mechanism, event boundary, or model condition that controls the answer, and end with a conclusion whose scope matches the evidence. This check is specific to case 7; do not replace it with a memorized generic sentence.
Worked case 8: Historical controls
Scenario. This year’s treated patients are compared with last year’s patients.
Reasoning. Changes in population, care standards, season, and measurement can confound the treatment comparison.
Case 8: Historical controls check: identify the exact statistical target in this scenario, verify the sign, denominator, assignment mechanism, event boundary, or model condition that controls the answer, and end with a conclusion whose scope matches the evidence. This check is specific to case 8; do not replace it with a memorized generic sentence.
Worked case 9: Matched observational groups
Scenario. Researchers match treated and untreated people on age and baseline score.
Reasoning. Matching can balance measured variables but does not guarantee balance on unmeasured confounders.
Case 9: Matched observational groups check: identify the exact statistical target in this scenario, verify the sign, denominator, assignment mechanism, event boundary, or model condition that controls the answer, and end with a conclusion whose scope matches the evidence. This check is specific to case 9; do not replace it with a memorized generic sentence.
Worked case 10: Causal caution
Scenario. A quasi-experimental estimate is large and statistically precise.
Reasoning. Precision does not erase design assumptions; the causal claim must still address how treatment assignment occurred and what confounding remains plausible.
Case 10: Causal caution check: identify the exact statistical target in this scenario, verify the sign, denominator, assignment mechanism, event boundary, or model condition that controls the answer, and end with a conclusion whose scope matches the evidence. This check is specific to case 10; do not replace it with a memorized generic sentence.
Worked case 11: Better comparison
Scenario. An untreated region with similar pre-intervention trends is added.
Reasoning. A credible comparison group strengthens the counterfactual: what likely would have happened to the treated region without intervention.
Case 11: Better comparison check: identify the exact statistical target in this scenario, verify the sign, denominator, assignment mechanism, event boundary, or model condition that controls the answer, and end with a conclusion whose scope matches the evidence. This check is specific to case 11; do not replace it with a memorized generic sentence.
Worked case 12: AP exam framing
Scenario. A prompt describes random treatment assignment.
Reasoning. Call it a randomized experiment and use the core AP rule: random assignment supports causal inference, while population scope depends separately on sampling.
Case 12: AP exam framing check: identify the exact statistical target in this scenario, verify the sign, denominator, assignment mechanism, event boundary, or model condition that controls the answer, and end with a conclusion whose scope matches the evidence. This check is specific to case 12; do not replace it with a memorized generic sentence.
Experimental vs Quasi-Experimental Design: Differences and Examples: 36 multiple-choice questions
These questions stay within this page’s topic. Work them after the concept and worked-case sections so practice reinforces the method rather than replacing instruction.
Question 1. Experimental vs. Quasi-Experimental Design
A city recreation department in Pine Ridge during a two-month observation window compares program satisfaction for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: D
At this city recreation department in Pine Ridge during a two-month observation window, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in program satisfaction, so the result supports an association for these groups but not a strong causal conclusion.
Question 2. Experimental vs. Quasi-Experimental Design
A grocery cooperative in Desert County during a follow-up evaluation period compares checkout time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: B
At this grocery cooperative in Desert County during a follow-up evaluation period, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in checkout time, so the result supports an association for these groups but not a strong causal conclusion.
Question 3. Experimental vs. Quasi-Experimental Design
A community bank in Coastal Plains during a semester-long cohort study compares mobile-deposit adoption for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: C
At this community bank in Coastal Plains during a semester-long cohort study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in mobile-deposit adoption, so the result supports an association for these groups but not a strong causal conclusion.
Question 4. Experimental vs. Quasi-Experimental Design
A housing authority in New England network during a monthly quality review compares application processing time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: A
At this housing authority in New England network during a monthly quality review, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in application processing time, so the result supports an association for these groups but not a strong causal conclusion.
Question 5. Experimental vs. Quasi-Experimental Design
A regional airport authority in Coastal Plains during a two-month observation window compares security wait time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: C
At this regional airport authority in Coastal Plains during a two-month observation window, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in security wait time, so the result supports an association for these groups but not a strong causal conclusion.
Question 6. Experimental vs. Quasi-Experimental Design
A community bank in Mountain Region during a weekday operations study compares mobile-deposit adoption for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: D
At this community bank in Mountain Region during a weekday operations study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in mobile-deposit adoption, so the result supports an association for these groups but not a strong causal conclusion.
Question 7. Experimental vs. Quasi-Experimental Design
A public high school in Prairie District during a fall 2026 audit compares algebra benchmark completion for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: C
At this public high school in Prairie District during a fall 2026 audit, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in algebra benchmark completion, so the result supports an association for these groups but not a strong causal conclusion.
Question 8. Experimental vs. Quasi-Experimental Design
A county library in South Harbor during a six-week field trial compares weekly program attendance for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: A
At this county library in South Harbor during a six-week field trial, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in weekly program attendance, so the result supports an association for these groups but not a strong causal conclusion.
Question 9. Experimental vs. Quasi-Experimental Design
A regional airport authority in Riverbend during a randomized pilot period compares security wait time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: C
At this regional airport authority in Riverbend during a randomized pilot period, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in security wait time, so the result supports an association for these groups but not a strong causal conclusion.
Question 10. Experimental vs. Quasi-Experimental Design
A digital learning platform in New England network during a monthly quality review compares lesson completion for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: A
At this digital learning platform in New England network during a monthly quality review, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in lesson completion, so the result supports an association for these groups but not a strong causal conclusion.
Question 11. Experimental vs. Quasi-Experimental Design
A community college in Cedar Grove during a six-week field trial compares course completion for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: A
At this community college in Cedar Grove during a six-week field trial, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in course completion, so the result supports an association for these groups but not a strong causal conclusion.
Question 12. Experimental vs. Quasi-Experimental Design
A digital learning platform in Cedar Grove during a baseline measurement week compares lesson completion for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: D
At this digital learning platform in Cedar Grove during a baseline measurement week, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in lesson completion, so the result supports an association for these groups but not a strong causal conclusion.
Question 13. Experimental vs. Quasi-Experimental Design
A regional airport authority in Sunbelt district during a randomized pilot period compares security wait time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: A
At this regional airport authority in Sunbelt district during a randomized pilot period, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in security wait time, so the result supports an association for these groups but not a strong causal conclusion.
Question 14. Experimental vs. Quasi-Experimental Design
A solar installer in Cedar Grove during a quarterly performance study compares daily energy output for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: D
At this solar installer in Cedar Grove during a quarterly performance study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in daily energy output, so the result supports an association for these groups but not a strong causal conclusion.
Question 15. Experimental vs. Quasi-Experimental Design
A farm cooperative in Desert County during a yearly program evaluation compares crop yield for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: B
At this farm cooperative in Desert County during a yearly program evaluation, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in crop yield, so the result supports an association for these groups but not a strong causal conclusion.
Question 16. Experimental vs. Quasi-Experimental Design
A state park in New England network during a six-week field trial compares trail-use duration for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: D
At this state park in New England network during a six-week field trial, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in trail-use duration, so the result supports an association for these groups but not a strong causal conclusion.
Question 17. Experimental vs. Quasi-Experimental Design
A city transit agency in Central County during a multiweek validation study compares on-time arrival for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: B
At this city transit agency in Central County during a multiweek validation study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in on-time arrival, so the result supports an association for these groups but not a strong causal conclusion.
Question 18. Experimental vs. Quasi-Experimental Design
A regional hospital in Atlantic Corridor during a service-improvement study compares appointment completion for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: B
At this regional hospital in Atlantic Corridor during a service-improvement study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in appointment completion, so the result supports an association for these groups but not a strong causal conclusion.
Question 19. Experimental vs. Quasi-Experimental Design
A municipal emergency dispatch center in Sunbelt district during a quarterly performance study compares response time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: B
At this municipal emergency dispatch center in Sunbelt district during a quarterly performance study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in response time, so the result supports an association for these groups but not a strong causal conclusion.
Question 20. Experimental vs. Quasi-Experimental Design
A university advising center in Great Lakes during a fall 2026 audit compares appointment wait time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: B
At this university advising center in Great Lakes during a fall 2026 audit, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in appointment wait time, so the result supports an association for these groups but not a strong causal conclusion.
Question 21. Experimental vs. Quasi-Experimental Design
A food safety laboratory in Riverbend during a regional benchmarking study compares sample concentration for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: A
At this food safety laboratory in Riverbend during a regional benchmarking study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in sample concentration, so the result supports an association for these groups but not a strong causal conclusion.
Question 22. Experimental vs. Quasi-Experimental Design
A digital learning platform in Westview during a follow-up evaluation period compares lesson completion for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: B
At this digital learning platform in Westview during a follow-up evaluation period, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in lesson completion, so the result supports an association for these groups but not a strong causal conclusion.
Question 23. Experimental vs. Quasi-Experimental Design
A digital learning platform in Central County during a yearly program evaluation compares lesson completion for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: D
At this digital learning platform in Central County during a yearly program evaluation, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in lesson completion, so the result supports an association for these groups but not a strong causal conclusion.
Question 24. Experimental vs. Quasi-Experimental Design
A university advising center in Pacific Northwest during a baseline measurement week compares appointment wait time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: A
At this university advising center in Pacific Northwest during a baseline measurement week, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in appointment wait time, so the result supports an association for these groups but not a strong causal conclusion.
Question 25. Experimental vs. Quasi-Experimental Design
A wildlife clinic in Great Lakes during a six-week field trial compares recovery time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: B
At this wildlife clinic in Great Lakes during a six-week field trial, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in recovery time, so the result supports an association for these groups but not a strong causal conclusion.
Question 26. Experimental vs. Quasi-Experimental Design
A regional manufacturer in Midwest consortium during a service-improvement study compares part diameter for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: D
At this regional manufacturer in Midwest consortium during a service-improvement study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in part diameter, so the result supports an association for these groups but not a strong causal conclusion.
Question 27. Experimental vs. Quasi-Experimental Design
A county library in Sunbelt district during a community outreach cycle compares weekly program attendance for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: B
At this county library in Sunbelt district during a community outreach cycle, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in weekly program attendance, so the result supports an association for these groups but not a strong causal conclusion.
Question 28. Experimental vs. Quasi-Experimental Design
A digital learning platform in Westview during a service-improvement study compares lesson completion for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: B
At this digital learning platform in Westview during a service-improvement study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in lesson completion, so the result supports an association for these groups but not a strong causal conclusion.
Question 29. Experimental vs. Quasi-Experimental Design
A municipal water office in Lakeside district during a yearly program evaluation compares monthly household use for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: D
At this municipal water office in Lakeside district during a yearly program evaluation, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in monthly household use, so the result supports an association for these groups but not a strong causal conclusion.
Question 30. Experimental vs. Quasi-Experimental Design
A municipal water office in Prairie District during a semester-long cohort study compares monthly household use for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: B
At this municipal water office in Prairie District during a semester-long cohort study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in monthly household use, so the result supports an association for these groups but not a strong causal conclusion.
Question 31. Experimental vs. Quasi-Experimental Design
A university advising center in Prairie District during a winter readiness review compares appointment wait time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: B
At this university advising center in Prairie District during a winter readiness review, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in appointment wait time, so the result supports an association for these groups but not a strong causal conclusion.
Question 32. Experimental vs. Quasi-Experimental Design
A municipal emergency dispatch center in Lakeside district during a service-improvement study compares response time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: B
At this municipal emergency dispatch center in Lakeside district during a service-improvement study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in response time, so the result supports an association for these groups but not a strong causal conclusion.
Question 33. Experimental vs. Quasi-Experimental Design
A regional hospital in Midwest consortium during a winter readiness review compares appointment completion for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: A
At this regional hospital in Midwest consortium during a winter readiness review, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in appointment completion, so the result supports an association for these groups but not a strong causal conclusion.
Question 34. Experimental vs. Quasi-Experimental Design
A university advising center in Metro East during a weekday operations study compares appointment wait time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: C
At this university advising center in Metro East during a weekday operations study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in appointment wait time, so the result supports an association for these groups but not a strong causal conclusion.
Question 35. Experimental vs. Quasi-Experimental Design
A digital learning platform in Metro East during a summer implementation review compares lesson completion for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: C
At this digital learning platform in Metro East during a summer implementation review, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in lesson completion, so the result supports an association for these groups but not a strong causal conclusion.
Question 36. Experimental vs. Quasi-Experimental Design
A community bank in Mountain Region during a quarterly performance study compares mobile-deposit adoption for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
Answer: D
At this community bank in Mountain Region during a quarterly performance study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in mobile-deposit adoption, so the result supports an association for these groups but not a strong causal conclusion.
Experimental vs Quasi-Experimental Design: Differences and Examples: 14 free-response questions
For each response, show the statistical reasoning, use the scenario’s language, and state only the conclusion supported by the design or probability model.
FRQ set 1: Experimental vs. Quasi-Experimental Design
Scenario. A regional airport authority in Lakeside district during a follow-up evaluation period compares security wait time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
- Identify the observational units, population, treatments or sampling frame, and the design goal.
- Describe the selection or assignment mechanism precisely enough that another researcher could implement it.
- Explain how the design controls bias, confounding, or variability, and identify a remaining limitation.
- State the scope of inference: population generalization, causal conclusion, both, or neither.
Model response
At this regional airport authority in Lakeside district during a follow-up evaluation period, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in security wait time, so the result supports an association for these groups but not a strong causal conclusion.
FRQ set 2: Experimental vs. Quasi-Experimental Design
Scenario. A city transit agency in Cedar Grove during a summer implementation review compares on-time arrival for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
- Identify the observational units, population, treatments or sampling frame, and the design goal.
- Describe the selection or assignment mechanism precisely enough that another researcher could implement it.
- Explain how the design controls bias, confounding, or variability, and identify a remaining limitation.
- State the scope of inference: population generalization, causal conclusion, both, or neither.
Model response
At this city transit agency in Cedar Grove during a summer implementation review, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in on-time arrival, so the result supports an association for these groups but not a strong causal conclusion.
FRQ set 3: Experimental vs. Quasi-Experimental Design
Scenario. A state park in Pacific Northwest during a yearly program evaluation compares trail-use duration for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
- Identify the observational units, population, treatments or sampling frame, and the design goal.
- Describe the selection or assignment mechanism precisely enough that another researcher could implement it.
- Explain how the design controls bias, confounding, or variability, and identify a remaining limitation.
- State the scope of inference: population generalization, causal conclusion, both, or neither.
Model response
At this state park in Pacific Northwest during a yearly program evaluation, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in trail-use duration, so the result supports an association for these groups but not a strong causal conclusion.
FRQ set 4: Experimental vs. Quasi-Experimental Design
Scenario. A community college in North Valley during a service-improvement study compares course completion for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
- Identify the observational units, population, treatments or sampling frame, and the design goal.
- Describe the selection or assignment mechanism precisely enough that another researcher could implement it.
- Explain how the design controls bias, confounding, or variability, and identify a remaining limitation.
- State the scope of inference: population generalization, causal conclusion, both, or neither.
Model response
At this community college in North Valley during a service-improvement study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in course completion, so the result supports an association for these groups but not a strong causal conclusion.
FRQ set 5: Experimental vs. Quasi-Experimental Design
Scenario. A food safety laboratory in Westview during a spring 2027 pilot compares sample concentration for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
- Identify the observational units, population, treatments or sampling frame, and the design goal.
- Describe the selection or assignment mechanism precisely enough that another researcher could implement it.
- Explain how the design controls bias, confounding, or variability, and identify a remaining limitation.
- State the scope of inference: population generalization, causal conclusion, both, or neither.
Model response
At this food safety laboratory in Westview during a spring 2027 pilot, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in sample concentration, so the result supports an association for these groups but not a strong causal conclusion.
FRQ set 6: Experimental vs. Quasi-Experimental Design
Scenario. A digital learning platform in Pacific Northwest during a spring 2027 pilot compares lesson completion for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
- Identify the observational units, population, treatments or sampling frame, and the design goal.
- Describe the selection or assignment mechanism precisely enough that another researcher could implement it.
- Explain how the design controls bias, confounding, or variability, and identify a remaining limitation.
- State the scope of inference: population generalization, causal conclusion, both, or neither.
Model response
At this digital learning platform in Pacific Northwest during a spring 2027 pilot, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in lesson completion, so the result supports an association for these groups but not a strong causal conclusion.
FRQ set 7: Experimental vs. Quasi-Experimental Design
Scenario. A community bank in Capital Region during a school-year data collection compares mobile-deposit adoption for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
- Identify the observational units, population, treatments or sampling frame, and the design goal.
- Describe the selection or assignment mechanism precisely enough that another researcher could implement it.
- Explain how the design controls bias, confounding, or variability, and identify a remaining limitation.
- State the scope of inference: population generalization, causal conclusion, both, or neither.
Model response
At this community bank in Capital Region during a school-year data collection, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in mobile-deposit adoption, so the result supports an association for these groups but not a strong causal conclusion.
FRQ set 8: Experimental vs. Quasi-Experimental Design
Scenario. A grocery cooperative in Pacific Northwest during a multiweek validation study compares checkout time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
- Identify the observational units, population, treatments or sampling frame, and the design goal.
- Describe the selection or assignment mechanism precisely enough that another researcher could implement it.
- Explain how the design controls bias, confounding, or variability, and identify a remaining limitation.
- State the scope of inference: population generalization, causal conclusion, both, or neither.
Model response
At this grocery cooperative in Pacific Northwest during a multiweek validation study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in checkout time, so the result supports an association for these groups but not a strong causal conclusion.
FRQ set 9: Experimental vs. Quasi-Experimental Design
Scenario. A community bank in Central County during a service-improvement study compares mobile-deposit adoption for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
- Identify the observational units, population, treatments or sampling frame, and the design goal.
- Describe the selection or assignment mechanism precisely enough that another researcher could implement it.
- Explain how the design controls bias, confounding, or variability, and identify a remaining limitation.
- State the scope of inference: population generalization, causal conclusion, both, or neither.
Model response
At this community bank in Central County during a service-improvement study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in mobile-deposit adoption, so the result supports an association for these groups but not a strong causal conclusion.
FRQ set 10: Experimental vs. Quasi-Experimental Design
Scenario. A regional manufacturer in Mountain Region during a follow-up evaluation period compares part diameter for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
- Identify the observational units, population, treatments or sampling frame, and the design goal.
- Describe the selection or assignment mechanism precisely enough that another researcher could implement it.
- Explain how the design controls bias, confounding, or variability, and identify a remaining limitation.
- State the scope of inference: population generalization, causal conclusion, both, or neither.
Model response
At this regional manufacturer in Mountain Region during a follow-up evaluation period, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in part diameter, so the result supports an association for these groups but not a strong causal conclusion.
FRQ set 11: Experimental vs. Quasi-Experimental Design
Scenario. A public health department in Desert County during a winter readiness review compares vaccination appointment completion for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
- Identify the observational units, population, treatments or sampling frame, and the design goal.
- Describe the selection or assignment mechanism precisely enough that another researcher could implement it.
- Explain how the design controls bias, confounding, or variability, and identify a remaining limitation.
- State the scope of inference: population generalization, causal conclusion, both, or neither.
Model response
At this public health department in Desert County during a winter readiness review, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in vaccination appointment completion, so the result supports an association for these groups but not a strong causal conclusion.
FRQ set 12: Experimental vs. Quasi-Experimental Design
Scenario. A wildlife clinic in Coastal Plains during a service-improvement study compares recovery time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
- Identify the observational units, population, treatments or sampling frame, and the design goal.
- Describe the selection or assignment mechanism precisely enough that another researcher could implement it.
- Explain how the design controls bias, confounding, or variability, and identify a remaining limitation.
- State the scope of inference: population generalization, causal conclusion, both, or neither.
Model response
At this wildlife clinic in Coastal Plains during a service-improvement study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in recovery time, so the result supports an association for these groups but not a strong causal conclusion.
FRQ set 13: Experimental vs. Quasi-Experimental Design
Scenario. A wildlife clinic in Capital Region during a weekday operations study compares recovery time for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
- Identify the observational units, population, treatments or sampling frame, and the design goal.
- Describe the selection or assignment mechanism precisely enough that another researcher could implement it.
- Explain how the design controls bias, confounding, or variability, and identify a remaining limitation.
- State the scope of inference: population generalization, causal conclusion, both, or neither.
Model response
At this wildlife clinic in Capital Region during a weekday operations study, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in recovery time, so the result supports an association for these groups but not a strong causal conclusion.
FRQ set 14: Experimental vs. Quasi-Experimental Design
Scenario. A public health department in Central County during a baseline measurement week compares vaccination appointment completion for people who voluntarily chose Program A with those who voluntarily chose Program B. Can the difference be interpreted causally?
- Identify the observational units, population, treatments or sampling frame, and the design goal.
- Describe the selection or assignment mechanism precisely enough that another researcher could implement it.
- Explain how the design controls bias, confounding, or variability, and identify a remaining limitation.
- State the scope of inference: population generalization, causal conclusion, both, or neither.
Model response
At this public health department in Central County during a baseline measurement week, participants chose programs rather than being randomly assigned. Self-selection and other confounders can explain the observed difference in vaccination appointment completion, so the result supports an association for these groups but not a strong causal conclusion.
Continue with a connected AP Statistics skill
Use the next page to connect this method to a neighboring concept rather than repeating the same calculation pattern.
