CPE Evaluation Methods & Reporting 5 — Questions and Answers
Question 1: Which design provides the strongest evidence of causal attribution in program evaluation?
- Pre-post single group design
- Matched comparison group design
- Randomized controlled trial (Correct answer)
- Interrupted time series without control group
Correct answer: Randomized controlled trial
Randomized controlled trials (RCTs) use random assignment to create equivalent groups, providing the strongest control for selection bias and confounding variables.
Question 2: Regression discontinuity design (RDD) is appropriate when:
- Program participants are assigned based on a continuous cutoff score (Correct answer)
- Two groups are perfectly matched on all demographic variables
- The program uses a lottery to determine participant selection
- Outcomes are measured at three or more time points
Correct answer: Program participants are assigned based on a continuous cutoff score
RDD exploits a cutoff score (e.g., income threshold, test score) used for program assignment to estimate causal effects at the threshold.
Question 3: Social Return on Investment (SROI) differs from traditional cost-benefit analysis primarily because SROI:
- Uses only government-approved economic data sources
- Monetizes social, environmental, and economic value using stakeholder-defined proxies (Correct answer)
- Excludes unintended consequences from calculations
- Requires a randomized control group for validity
Correct answer: Monetizes social, environmental, and economic value using stakeholder-defined proxies
SROI assigns monetary values to social and environmental outcomes using stakeholder-defined proxies, broadening the scope beyond traditional economic metrics.
Question 4: An evaluator conducting focus groups notices one dominant participant repeatedly silencing others. The BEST immediate response is to:
- End the focus group and replace it with individual interviews
- Use structured turn-taking techniques to ensure all voices are heard (Correct answer)
- Remove the dominant participant from the group
- Note the dynamic in field notes and continue without intervention
Correct answer: Use structured turn-taking techniques to ensure all voices are heard
Using facilitation techniques like structured turn-taking or direct invitation addresses dominance dynamics while preserving the group discussion.
Question 5: The AEA Program Evaluation Standards include which of the following categories?
- Validity, reliability, generalizability, and significance
- Utility, feasibility, propriety, accuracy, and evaluation accountability (Correct answer)
- Formative, summative, developmental, and empowerment standards
- Quantitative, qualitative, mixed-methods, and economic standards
Correct answer: Utility, feasibility, propriety, accuracy, and evaluation accountability
The Joint Committee's Program Evaluation Standards include five categories: utility, feasibility, propriety, accuracy, and evaluation accountability.
Question 6: When an evaluator discovers mid-study that the program significantly changed its intervention model, the MOST appropriate action is to:
- Continue the original evaluation plan to maintain research integrity
- Terminate the evaluation and return all funding
- Document the change, revise the evaluation questions, and renegotiate the evaluation plan with stakeholders (Correct answer)
- Analyze only data collected before the program change
Correct answer: Document the change, revise the evaluation questions, and renegotiate the evaluation plan with stakeholders
Mid-program changes require the evaluator to adapt the evaluation plan and renegotiate with stakeholders to ensure the evaluation remains relevant and valid.
Question 7: Which technique is used to synthesize findings across multiple evaluations of similar programs to draw broader conclusions?
- Case study analysis
- Meta-analysis or systematic review (Correct answer)
- Cross-site evaluation
- Realist synthesis
Correct answer: Meta-analysis or systematic review
Meta-analysis statistically pools effect sizes across multiple studies, and systematic reviews synthesize evidence, both producing generalizable conclusions about program effectiveness.
Which design provides the strongest evidence of causal attribution in program evaluation?