CPE Data Collection & Analysis 4 — Questions and Answers
Question 1: When designing a data collection instrument for a diverse population, an evaluator should prioritize:
- Using technical jargon to appear credible
- Cultural responsiveness and plain language (Correct answer)
- Maximizing the number of questions
- Avoiding any demographic questions
Correct answer: Cultural responsiveness and plain language
Culturally responsive instruments in plain language ensure all participants can understand and respond accurately.
Question 2: A program evaluation finds a Pearson correlation of r = 0.85 between two variables. This indicates:
- One variable causes the other
- A strong positive linear relationship (Correct answer)
- A weak association
- No meaningful relationship
Correct answer: A strong positive linear relationship
A Pearson r of 0.85 indicates a strong positive linear relationship, though causation cannot be inferred from correlation alone.
Question 3: An evaluator wants to track changes in program outcomes over multiple time points without a comparison group. This design is called a:
- Randomized controlled trial
- Single-group time-series design (Correct answer)
- Matched comparison design
- Cross-sectional survey
Correct answer: Single-group time-series design
A single-group time-series design measures the same group at multiple time points to assess change over time without a comparison group.
Question 4: Which measure of data spread indicates how much individual scores deviate from the group average?
- Mean
- Median
- Standard deviation (Correct answer)
- Frequency
Correct answer: Standard deviation
Standard deviation quantifies the average distance of individual scores from the mean, indicating data spread.
Question 5: A CPE evaluator administers a survey to all 500 program participants rather than a sample. This is called:
- Stratified sampling
- A census (Correct answer)
- Cluster sampling
- Systematic sampling
Correct answer: A census
Collecting data from every member of the population rather than a subset is called a census.
Question 6: Interrater reliability is most important when data collection involves:
- Automated software scoring
- Multiple human coders categorizing qualitative data (Correct answer)
- Single-item surveys
- Administrative record extraction
Correct answer: Multiple human coders categorizing qualitative data
Interrater reliability ensures that multiple human coders assign the same categories to data consistently, reducing subjective bias.
Question 7: Which ethical principle requires evaluators to protect participants from harm during data collection?
- Justice
- Autonomy
- Beneficence (Correct answer)
- Fidelity
Correct answer: Beneficence
Beneficence obligates evaluators to maximize benefits and minimize harm to participants throughout the evaluation process.
CPE Data Collection & Analysis 4
When designing a data collection instrument for a diverse population, an evaluator should prioritize: