MSW Research and Program Evaluation 2 — Questions and Answers
Question 1: A social worker wants to measure whether a job-training program reduces unemployment. Which evaluation design best establishes causality?
- Pre-test/post-test with a control group (Correct answer)
- Post-test only with no comparison group
- Case study of three successful participants
- Focus group interviews with graduates
Correct answer: Pre-test/post-test with a control group
A pre-test/post-test design with a control group controls for threats to internal validity, allowing causal inferences.
Question 2: Which sampling method is most appropriate when no complete list of the target population exists?
- Snowball sampling (Correct answer)
- Simple random sampling
- Systematic sampling
- Stratified random sampling
Correct answer: Snowball sampling
Snowball sampling uses referrals from known participants to reach hidden or hard-to-find populations.
Question 3: A researcher finds a p-value of 0.03. What does this indicate?
- There is a 3% probability the results occurred by chance under the null hypothesis (Correct answer)
- The effect size is small
- The study has 97% power
- The results are clinically significant
Correct answer: There is a 3% probability the results occurred by chance under the null hypothesis
A p-value represents the probability of obtaining results as extreme as observed if the null hypothesis were true.
Question 4: In program evaluation, 'process evaluation' primarily examines:
- Whether the program was implemented as intended (Correct answer)
- Long-term client outcomes after discharge
- Cost-effectiveness of the intervention
- Statistical significance of behavior changes
Correct answer: Whether the program was implemented as intended
Process evaluation (also called implementation evaluation) assesses fidelity to the program model and service delivery.
Question 5: A social work researcher uses an unvalidated instrument to measure depression. This most directly threatens which aspect of the study?
- Measurement validity (Correct answer)
- External validity
- Statistical power
- Sampling adequacy
Correct answer: Measurement validity
Using an unvalidated instrument threatens measurement validity because it may not accurately capture the construct being studied.
Question 6: Which research approach is most consistent with an interpretivist paradigm?
- Grounded theory (Correct answer)
- Randomized controlled trial
- Meta-analysis
- Survey with Likert scales
Correct answer: Grounded theory
Grounded theory is a qualitative method that seeks to generate theory from data, consistent with interpretivist assumptions about subjective meaning.
Question 7: When evaluating a domestic violence shelter program, an evaluator compares outcomes of women who sought services versus those who did not. The biggest threat to internal validity is:
- Selection bias (Correct answer)
- Instrumentation error
- Testing effect
- Statistical regression
Correct answer: Selection bias
Selection bias occurs because women who self-select into shelter services likely differ in motivation and resources from those who do not seek services.
MSW Research and Program Evaluation 2
A social worker wants to measure whether a job-training program reduces unemployment.
Which evaluation design best establishes causality?