Page 11 of 13
CM6.1-6 | Biostatistics for Community Medicine — Graded Quiz
Click any question card to reveal the correct answer.
Parity (0, 1, 2, 3, ≥4) recorded as a count is classified as which type of data?
Correct. Parity is a whole-number count (0, 1, 2, …), making it discrete quantitative.
Discrete quantitative: countable whole numbers (parity, number of children). Continuous quantitative: any value on a scale (weight, blood pressure). Qualitative/categorical: no numerical meaning (nominal = no order; ordinal = ordered categories).
Incorrect. Parity is a count of pregnancies that can only take whole-number values → discrete quantitative.
Click to reveal answer
In a perfectly symmetrical, bell-shaped (normal) distribution, which of the following is TRUE?
Correct. In a perfect normal distribution, mean = median = mode — all three coincide at the centre of symmetry.
Relationship with skew: Normal: Mean=Median=Mode. Positive skew (right tail): Mean>Median>Mode. Negative skew (left tail): Mean<Median<Mode. Memory aid: in positive skew, the 'mean is pulled toward the long right tail'.
Incorrect. The three measures of central tendency coincide at the centre in a perfectly symmetrical normal distribution.
Click to reveal answer
A researcher divides a district population into urban and rural strata, then randomly selects 200 individuals from each stratum. This is:
Correct. Dividing the population into subgroups (strata) and sampling randomly within each stratum is stratified random sampling.
Key distinction from cluster sampling: in stratified, you sample from WITHIN each stratum (subgroups are heterogeneous units); in cluster, you randomly select whole clusters and study everyone in them. Stratified improves precision for subgroup estimates.
Incorrect. Stratified random sampling involves: (1) dividing the population into mutually exclusive strata based on a key variable, then (2) performing random sampling within each stratum.
Click to reveal answer
Before conducting a study comparing mean systolic blood pressure (SBP) between smokers and non-smokers, a researcher states: 'There is no difference in mean SBP between smokers and non-smokers.' This statement is the:
Correct. The null hypothesis (H₀) always states 'no difference' or 'no association'. Statistical tests seek to reject H₀.
H₀ = null hypothesis: the default position of 'no effect/difference'. H₁ = alternative hypothesis: the research question (there IS an effect/difference). The statistical test generates a p-value: if p < α, we reject H₀ in favour of H₁. We never 'accept' H₀; we either reject it or fail to reject it.
Incorrect. A statement of no difference is the null hypothesis (H₀). The alternative hypothesis (H₁) asserts that a difference exists.
Click to reveal answer
Setting α = 0.05 in a hypothesis test means the investigator accepts a maximum probability of ___% of making a Type I error.
Correct. α = 0.05 means a 5% probability of incorrectly rejecting a true null hypothesis (Type I error = false positive).
Error types: Type I (α, false positive): rejecting H₀ when it is actually true. Type II (β, false negative): failing to reject H₀ when it is actually false. Power = 1 − β. Common values: α = 0.05 (allows 5% Type I error), β = 0.20 (allows 20% Type II error → power = 80%).
Incorrect. α is the significance level and directly equals the pre-set probability of a Type I error.
Click to reveal answer
A nutritionist measures haemoglobin (g/dL) in 30 women before and after a 3-month iron supplementation programme. Haemoglobin differences are normally distributed. The MOST appropriate test is:
Correct. Pre-post measurements on the SAME individuals → paired data. Normal distribution of differences → paired t-test.
Paired t-test: tests whether the mean of paired differences (after−before) is significantly different from zero. Applicable: same subject before and after, or matched pairs. Requirement: differences (not the raw values) must be normally distributed. Non-parametric equivalent: Wilcoxon signed-rank test.
Incorrect. Same subjects measured before and after = paired/dependent data. Paired t-test analyses the differences within individuals, reducing inter-individual variability.
Click to reveal answer
A study finds the mean diastolic BP difference between treated and untreated hypertensives is −8 mmHg (95% CI: −12 to −4 mmHg). Which conclusion is CORRECT?
Correct. The 95% CI (−12 to −4) does NOT include zero, indicating the reduction is statistically significant (the null value of 0 difference is excluded).
95% CI interpretation: if the CI for a difference excludes zero, the result is statistically significant at the 5% level. Here, −12 to −4 entirely below zero → the treatment clearly reduces diastolic BP. Width of CI reflects precision (narrow = precise, wide = imprecise) but width alone does not determine significance — only whether null is included matters.
Incorrect. When a 95% CI does NOT include the null value (zero for a difference), the result is statistically significant at α = 0.05.
Click to reveal answer
If the variance of a dataset is 49 kg², the standard deviation is:
Correct. SD = √variance = √49 = 7 kg. SD is expressed in the same units as the original data.
Variance = Σ(x−x̄)²/(n−1); expressed in squared units. SD = √variance; expressed in original units — this is why SD is more intuitive for describing spread. A dataset with variance = 49 kg² has SD = 7 kg.
Incorrect. Standard deviation is the square root of variance. √49 = 7 kg.
Click to reveal answer
A community medicine resident enters data into SPSS and performs an independent samples t-test. The output shows p = 0.038. The pre-set significance level was α = 0.05. The correct interpretation is:
Correct. p = 0.038 < α = 0.05 → reject H₀ → the difference is statistically significant at the 5% level.
Decision rule: if p < α, reject H₀ (result is statistically significant). If p ≥ α, fail to reject H₀ (result is not statistically significant). Note: we NEVER 'accept' H₀ — we only fail to reject it. Understanding software output is a core CM6.5 competency.
Incorrect. When p < α, we reject the null hypothesis. Since 0.038 < 0.05, H₀ is rejected.
Click to reveal answer
A medical officer records the following ages (years) of 7 patients with typhoid fever at a PHC: 4, 6, 7, 8, 8, 10, 45. The BEST summary of central tendency and spread is:
Correct. With an outlier (45 yr — an adult among children), the median (8 yr) and IQR are the robust, appropriate summary statistics.
When data are skewed or contain outliers: use Median + IQR. When data are normally distributed: use Mean ± SD. IQR = Q3−Q1. For these 7 values sorted: 4,6,7,8,8,10,45; Median = 4th value = 8; Q1 = median of lower half (4,6,7) = 6; Q3 = median of upper half (8,10,45) = 10; IQR = 10−6 = 4 yr.
Incorrect. The outlier (45 yr) inflates the mean to 12.6 yr (a misleading estimate of the 'typical' patient). Median = 8 yr and IQR = 10−6 = 4 yr are unaffected by the outlier.
Click to reveal answer