The variation in a data set is compared with that of another data set using?

Explanation

## Correct Answer: D. Coefficient of variation When comparing variability across two or more datasets with different means or different units of measurement, **coefficient of variation (CV)** is the only appropriate measure. CV is a standardized, dimensionless measure calculated as (SD/Mean) × 100, expressed as a percentage. This normalization allows direct comparison of relative variability regardless of scale. For example, comparing blood glucose variability (mg/dL) with hemoglobin variability (g/dL) in the same patient population requires CV because the absolute standard deviations are on different scales. In Indian clinical epidemiology, CV is essential when comparing disease burden across populations with different baseline characteristics—such as comparing nutritional status variation in children from different socioeconomic strata or comparing anthropometric measurements across age groups. Standard deviation and variance are absolute measures tied to the original scale and cannot be meaningfully compared across datasets with different units or means. Standard error of mean measures precision of the sample mean estimate, not dataset variability itself. ## Why the other options are wrong **A. Standard deviation** — SD is an absolute measure of dispersion tied to the original scale and units. It cannot be used to compare variability between datasets with different means or different measurement units. For instance, SD of blood pressure (mmHg) cannot be directly compared with SD of heart rate (bpm) without standardization. This is the NBE trap—students confuse 'variability measure' with 'comparative variability measure.' **B. Standard error of mean** — SEM measures the precision or reliability of the sample mean as an estimate of the population mean, not the variability within the dataset itself. SEM = SD/√n, so it decreases with larger sample size even if dataset variability remains constant. SEM is used for confidence intervals and hypothesis testing, not for comparing inherent variability between two datasets. **C. Variance** — Variance (SD²) is also an absolute measure of dispersion in squared units, making it even less interpretable than SD when comparing across datasets. Like SD, variance cannot account for differences in scale or mean magnitude. A dataset with higher variance may actually have lower relative variability if its mean is proportionally larger. ## High-Yield Facts - **Coefficient of variation (CV)** = (SD/Mean) × 100 is a dimensionless, standardized measure used to compare relative variability across datasets with different means or units. - CV is essential in Indian public health surveillance—comparing disease incidence variation across states with different population sizes or comparing nutritional indicators across age groups. - **CV > 30%** generally indicates high relative variability; **CV < 15%** indicates low relative variability (rule of thumb in epidemiological studies). - SD and variance are absolute measures and cannot be meaningfully compared across datasets with different scales; CV solves this by normalizing to the mean. - In clinical laboratories, CV is used to assess analytical precision of assays across different measurement ranges (e.g., glucose assay precision at 100 mg/dL vs. 300 mg/dL). ## Mnemonics **CV = Comparative Variability** When you need to **Compare** variability across different datasets or units, use **CV**. SD and Variance are absolute; CV is relative. **SCALE-FREE = CV** CV is **Scale-Free** (dimensionless, percentage-based), making it the only choice when comparing datasets with different means, units, or measurement ranges. ## NBE Trap NBE pairs "variation comparison" with "standard deviation" to trap students who confuse 'a measure of variability' with 'a measure for comparing variability across datasets.' The key discriminator is the phrase "compared with that of another data set"—this signals the need for a standardized, scale-independent measure. ## Clinical Pearl In Indian nutrition surveys (NFHS, CNNS), CV is routinely used to compare anthropometric variability across different age groups and socioeconomic strata. A child's weight SD cannot be directly compared with an adult's weight SD—but their CVs can be, revealing whether relative nutritional heterogeneity is higher in one group than another. _Reference: Park's Textbook of Preventive and Social Medicine, Ch. 10 (Biostatistics); Harrison's Principles of Internal Medicine, Ch. 5 (Quantitative Aspects of Clinical Reasoning)_