CPSS Applied Research Methods 4 — Questions and Answers
Question 1: A researcher uses Cohen's d = 1.2 to describe the difference in jump height between two training groups. How is this effect size classified?
- Small
- Medium
- Large (Correct answer)
- Trivial
Correct answer: Large
Cohen's d ≥ 0.8 is classified as a large effect size, and d = 1.2 substantially exceeds this threshold.
Question 2: In a single-subject research design, the A-B-A format refers to:
- Pre-test, intervention, post-test measured once
- Baseline, treatment, return-to-baseline phases (Correct answer)
- Alternating treatments in randomized order
- Three separate experimental groups
Correct answer: Baseline, treatment, return-to-baseline phases
The A-B-A design establishes a baseline (A), introduces treatment (B), then withdraws it to return to baseline (A) to confirm the intervention's effect.
Question 3: Which index is most appropriate for evaluating the agreement between two raters using ordinal data?
- Pearson r
- Intraclass correlation coefficient (ICC)
- Cohen's weighted kappa (Correct answer)
- Cronbach's alpha
Correct answer: Cohen's weighted kappa
Cohen's weighted kappa accounts for chance agreement and respects the ordered nature of ordinal rating scales.
Question 4: A sport scientist wants to identify factors that predict athletic injury risk over a season. Which statistical model is most appropriate?
- Paired t-test
- Simple linear regression
- Multiple logistic regression (Correct answer)
- One-way ANOVA
Correct answer: Multiple logistic regression
Multiple logistic regression models the probability of a binary outcome (injured vs. not injured) from multiple predictor variables.
Question 5: What does an intraclass correlation coefficient (ICC) of 0.95 indicate about a measurement tool?
- Poor reliability
- Moderate reliability
- Excellent reliability (Correct answer)
- Unacceptable validity
Correct answer: Excellent reliability
ICC values above 0.90 are generally classified as excellent reliability, indicating high consistency across repeated measurements.
Question 6: In a meta-analysis, a funnel plot is primarily used to assess:
- Heterogeneity across study designs
- Publication bias (Correct answer)
- Effect size magnitude
- Statistical power of included studies
Correct answer: Publication bias
An asymmetric funnel plot suggests that small studies with negative findings may be missing from the literature, indicating publication bias.
Question 7: Which condition must be met for a Pearson correlation coefficient to be a valid measure of association?
- Both variables are normally distributed and have a linear relationship (Correct answer)
- One variable is categorical and the other is continuous
- Data are collected on a nominal scale
- Sample size must exceed 200 participants
Correct answer: Both variables are normally distributed and have a linear relationship
Pearson r assumes both variables are continuous, normally distributed, and related in a linear fashion.
CPSS Applied Research Methods 4
A researcher uses Cohen's d = 1.2 to describe the difference in jump height between two training groups.
How is this effect size classified?