Correct Answer: D. Weak correlation
A correlation coefficient (r) quantifies the strength and direction of a linear relationship between two variables, ranging from −1 to +1. The magnitude of r—not its sign—determines strength. A coefficient less than 0.5 in absolute value indicates a weak correlation, meaning the two variables have a loose linear association. In Indian epidemiological studies (e.g., correlating BMI with blood pressure in community surveys), an r < 0.5 suggests that knowing one variable provides limited predictive power for the other. The interpretation scale is: |r| < 0.3 (very weak), 0.3–0.5 (weak), 0.5–0.7 (moderate), 0.7–0.9 (strong), and > 0.9 (very strong). Statistical significance (p-value) is a separate concept from correlation strength; a large sample can yield a statistically significant but weak correlation. Confidence intervals relate to estimation of population parameters, not correlation strength classification. Thus, r < 0.5 is definitionally classified as weak correlation.
Why the other options are wrong
A. Strong correlation — This is wrong because strong correlation typically requires |r| ≥ 0.7 or higher. An r < 0.5 is at the opposite end of the strength spectrum. This is a direct misinterpretation of the correlation coefficient scale and represents the most obvious trap for unprepared students. B. Confidence interval of 95% — This is wrong because confidence intervals and correlation coefficients are distinct statistical concepts. A 95% CI relates to the precision of an estimate (e.g., mean, proportion), not to the strength of correlation. This option conflates two unrelated biostatistical measures and tests whether students confuse different statistical tools. C. Not statistically significant — This is wrong because statistical significance (determined by p-value and sample size) is independent of correlation strength. A weak correlation (r < 0.5) can still be statistically significant in a large sample, and a strong correlation can be non-significant in a small sample. This trap confuses two orthogonal statistical concepts.
High-Yield Facts
- Correlation coefficient |r| < 0.3: very weak; 0.3–0.5: weak; 0.5–0.7: moderate; 0.7–0.9: strong; > 0.9: very strong
- Correlation strength is independent of statistical significance: large samples can yield significant weak correlations; small samples may have non-significant strong correlations
- Pearson's r measures only linear relationships: a non-linear but strong association may yield low r values
- Confidence intervals estimate population parameters (e.g., mean, difference in means), not correlation strength
- In Indian public health surveys, weak correlations (r < 0.5) between risk factors and outcomes often indicate multifactorial causation requiring multivariate analysis
Mnemonics
Correlation Strength Scale (0–1 Rule) 0–0.3 = Very weak; 0.3–0.5 = Weak; 0.5–0.7 = Moderate; 0.7–0.9 = Strong; 0.9–1.0 = Very strong. Use when interpreting any r value in epidemiological or clinical studies. SIPS: Significance ≠ Strength Significance (p-value) and Intensity (correlation strength) are Perfectly Separate. A large n can make weak r significant; small n can hide strong r. Prevents confusing statistical significance with effect size.
NBE Trap
NBE pairs "correlation coefficient" with "statistical significance" to trap students who conflate p-values with correlation strength. The question tests whether students know that r < 0.5 is a strength classification, not a significance statement.
Clinical Pearl
In Indian ICMR-led cohort studies (e.g., INTERHEART India), weak correlations between individual risk factors and myocardial infarction (r < 0.5) are common, prompting researchers to use logistic regression to identify independent predictors—a reminder that weak univariate correlation does not exclude clinical importance in multivariate models.
_Reference: Park's Textbook of Preventive and Social Medicine, Ch. 10 (Biostatistics); Harrison's Principles of Internal Medicine, Ch. 6 (Quantitative Aspects of Clinical Reasoning)_