HSP Research and Scientific Foundations 2 — Questions and Answers
Question 1: A researcher wants to determine if a new CBT protocol causes greater symptom reduction than treatment-as-usual. Which design best establishes causality?
- Cross-sectional survey with matched groups
- Randomized controlled trial with blinded outcome assessment (Correct answer)
- Prospective cohort study without randomization
- Case-control study comparing current and past patients
Correct answer: Randomized controlled trial with blinded outcome assessment
Randomized controlled trials with blinded assessment are the gold standard for establishing causal relationships between interventions and outcomes.
Question 2: In a factor analysis of a new psychological measure, a scree plot shows a clear break after the second eigenvalue. What does this suggest?
- The measure has poor construct validity
- Two factors best explain the variance in the data (Correct answer)
- The reliability coefficient is unacceptably low
- The sample size is insufficient for analysis
Correct answer: Two factors best explain the variance in the data
A scree plot break after the second eigenvalue indicates that two factors account for meaningful variance, supporting a two-factor solution.
Question 3: A health psychologist is reviewing a study reporting p = .049 for a primary outcome. Which statement best reflects current scientific standards?
- The result is definitively clinically significant
- Statistical significance alone does not confirm clinical meaningfulness (Correct answer)
- The finding should be immediately replicated in a meta-analysis
- A p-value just under .05 is more reliable than p = .001
Correct answer: Statistical significance alone does not confirm clinical meaningfulness
A p-value near the threshold indicates statistical significance but must be evaluated alongside effect size and clinical meaningfulness.
Question 4: Which type of validity is assessed when a new depression scale correlates strongly with a well-established depression measure?
- Content validity
- Discriminant validity
- Concurrent criterion validity (Correct answer)
- Face validity
Correct answer: Concurrent criterion validity
Concurrent criterion validity is demonstrated when a new measure correlates with an established gold-standard measure administered at the same time.
Question 5: A systematic review uses the GRADE framework. What does a 'low quality' GRADE rating primarily indicate?
- The review included fewer than five studies
- There is little confidence that the evidence reflects the true effect (Correct answer)
- The effect size is too small to be clinically meaningful
- The studies were not published in peer-reviewed journals
Correct answer: There is little confidence that the evidence reflects the true effect
A 'low' GRADE rating means further research is very likely to change confidence in the estimate of effect.
Question 6: A researcher uses snowball sampling to recruit participants with rare chronic conditions. The primary limitation of this approach is:
- It yields too large a sample for qualitative analysis
- Participants may share network characteristics, limiting generalizability (Correct answer)
- It violates APA ethical standards for research recruitment
- The sample cannot be used to calculate effect sizes
Correct answer: Participants may share network characteristics, limiting generalizability
Snowball sampling introduces network bias because recruits referred through social connections may share characteristics that limit generalizability.
Question 7: Cohen's d of 0.80 is conventionally interpreted as:
- Small effect
- Medium effect
- Large effect (Correct answer)
- Negligible effect
Correct answer: Large effect
By Cohen's conventions, d = 0.20 is small, d = 0.50 is medium, and d = 0.80 is large.
HSP Research and Scientific Foundations 2
A researcher wants to determine if a new CBT protocol causes greater symptom reduction than treatment-as-usual.
Which design best establishes causality?