BIOSTATISTICS • DESCRIPTIVE STATISTICS & VISUALIZATION

Mean, Median & Mode — Compute and interpret mean, median, mode

Three measures of central tendency that summarize where biological data cluster and guide clinical and research decisions.

Historical Context & Motivation

Long before modern biostatistics emerged as a formal discipline, scientists and administrators grappled with the challenge of summarizing large collections of numerical observations into single representative values. The need was intensely practical: governments required efficient ways to report population data, astronomers needed to reconcile discrepant measurements of celestial positions, and physicians sought to characterize 'normal' vital signs. The intellectual journey toward the mean, median, and mode spans several centuries and reflects evolving perspectives on what it means to capture the 'center' of a data distribution.

1669
Huygens and the Expected Value
Christiaan Huygens formalized the concept of expectation in his treatise on probability, laying the groundwork for the arithmetic mean as a probabilistic summary of repeated observations.
1756
Simpson's Advocacy for the Mean
Thomas Simpson demonstrated that the arithmetic mean of multiple observations reduces the influence of random errors, providing a formal justification for averaging in scientific measurement.
1774
Laplace and the Median
Pierre-Simon Laplace proposed the median as a measure that minimizes the sum of absolute deviations, recognizing its resistance to extreme values — a property now central to robust statistics.
1895
Pearson Formalizes the Mode
Karl Pearson introduced the term mode to describe the most frequently occurring value, and explored skewness relationships among the three measures as part of his broader system of frequency curves.
1920s–Present
Biostatistical Integration
With the rise of clinical trials and epidemiology, Fisher, Hill, and others embedded these measures into standard reporting frameworks. Today, the choice among mean, median, and mode directly influences clinical guidelines, drug approval decisions, and public health policy.

This history underscores a fundamental question that persists in modern biostatistical practice: which single number best represents the center of a dataset, and under what conditions does each measure yield the most faithful summary? The answer depends on the shape of the distribution, the type of variable, and the inferential goals of the analysis — all considerations we explore in the sections that follow.

Core Principles & Definitions

Measures of central tendency serve a singular purpose: to distill a collection of observations into one value that captures where data points congregate. In biostatistics, where datasets may range from a handful of patient lab values to genome-wide expression arrays containing millions of entries, choosing the appropriate measure is not merely a computational decision — it is an inferential one. Each of the three classical measures — mean, median, and mode — encodes different assumptions about the data-generating process and responds differently to distributional features such as skewness, outliers, and multimodality.

1

Arithmetic Mean (x̄)

The sum of all observed values divided by the number of observations. It incorporates every data point and acts as the balance point of the distribution — the value at which the sum of deviations equals zero. It is the most efficient estimator of the population center under normality but is sensitive to extreme values.
2

Median (x̃)

The value that divides an ordered dataset into two equal halves. Exactly 50% of observations fall at or below and 50% at or above. As a positional measure, it is unaffected by the magnitude of extreme values, making it the preferred summary for skewed distributions such as hospital length-of-stay or healthcare expenditures.
3

Mode

The value (or values) that occurs with the greatest frequency. It is the only measure of central tendency applicable to nominal data (e.g., blood type, genotype). Continuous data may have no exact repeating values, so the mode is often estimated from a frequency histogram or kernel density estimate.
4

Sensitivity to Outliers

The mean shifts toward outliers because it is a function of every observed value. The median shifts only if the number of values on each side changes. The mode is entirely unaffected. This hierarchy of robustness drives practical guidelines in clinical reporting.
5

Relationship Under Symmetry & Skew

In a perfectly symmetric, unimodal distribution the three measures coincide. Right (positive) skew pulls the mean above the median. Left (negative) skew pulls the mean below. Comparing the mean and median therefore serves as a quick diagnostic for skewness.
KEY TAKEAWAY
Think of the three measures as different 'cameras' focused on the same crowd. The mean is a drone shot that weights every person's position equally — a single person standing far from the group pulls the camera noticeably. The median is a ground-level photo from the person standing in the exact middle of the line — it doesn't matter how far the outliers wander. The mode is a heat-map view highlighting where the crowd is densest. Each perspective reveals something the others miss.

Visual Explanation — Distribution Shape and the Three Measures

In a symmetric distribution (left panel), all three measures coincide. Under right (positive) skew (center panel), the mean is pulled toward the long right tail, positioning it to the right of the median and mode. Under left (negative) skew (right panel), the pattern reverses. The ordering Mode–Median–Mean (or its reverse) serves as a rapid visual diagnostic for skewness in biostatistical data.

The diagram above illustrates the fundamental geometric relationship among the three measures. Biostatistical datasets are frequently right-skewed — consider hospital costs, viral load counts, or time to disease recurrence — meaning that the mean will systematically overstate the 'typical' patient experience relative to the median. In clinical reporting, regulatory agencies such as the FDA therefore often request both the mean and median to ensure transparency, particularly when evaluating endpoints like overall survival or healthcare utilization. Recognizing the skewness signature at a glance, simply by noting how far the mean sits from the median, is one of the most useful quick-check skills in applied biostatistics.

Mathematical Framework

Each measure of central tendency can be defined formally, enabling rigorous derivation of its statistical properties. Below we present the formulas for sample computations — the versions most commonly applied in biostatistical practice — alongside key mathematical properties that motivate their use.

ARITHMETIC MEAN
x̄ = (1/n) × Σᵢ₌₁ⁿ xᵢ
where is the sample mean, n is the number of observations, and xᵢ is the i-th observed value. The mean minimizes the sum of squared deviations: Σ(xᵢ − c)² is minimized when c = x̄.
MEDIAN — ODD n
x̃ = x₍₍ₙ₊₁₎/₂₎ when n is odd
Sort the data in ascending order. The median is the observation at position (n + 1)/2. For even n, the median is the average of the two central values: x̃ = (x₍ₙ/₂₎ + x₍ₙ/₂₊₁₎) / 2. The median minimizes the sum of absolute deviations: Σ|xᵢ − c| is minimized when c = x̃.
MODE
Mode = argmax f(x)
The mode is the value (or values) of x that maximizes the frequency function f(x). For grouped or continuous data, the mode is often estimated from the peak of a fitted density. A distribution with two modes is bimodal; three or more modes is multimodal.
📐 Pearson's Skewness Approximation
Karl Pearson proposed an empirical relationship linking the three measures in moderately skewed, unimodal distributions: Mean − Mode ≈ 3 × (Mean − Median). While this is an approximation and can break down for strongly skewed or multimodal data, it provides a useful rule of thumb in exploratory biostatistical analysis. If you know any two of the three measures, you can estimate the third.

From an estimation-theoretic perspective, the mean is the maximum likelihood estimator (MLE) of the population center when the underlying distribution is normal. Its variance equals σ²/n, meaning it becomes more precise as the sample size grows — a property formalized by the law of large numbers and the central limit theorem. The median, by contrast, has an asymptotic variance of π/(2n) × σ² ≈ 1.57σ²/n under normality, making it roughly 57% less efficient than the mean for Gaussian data. However, for heavy-tailed distributions common in biological data (e.g., lognormal drug concentrations), the median can actually be more efficient than the mean — a fact that underscores why the choice between these measures is fundamentally distribution-dependent.

Choosing the Right Measure — A Decision Framework

Selecting the appropriate measure of central tendency is one of the first decisions in any biostatistical analysis, and the choice is governed by three interrelated factors: the measurement scale (nominal, ordinal, interval, ratio), the distributional shape (symmetric versus skewed), and the presence of outliers or censored observations. The flowchart below codifies these decision rules into a practical algorithm.

This decision tree guides the choice of central tendency measure based on measurement scale, distributional shape, and outlier presence. For nominal data, the mode is the sole option. For ordinal data, the median is preferred. For interval/ratio data, symmetry and outlier considerations determine whether the mean or median is the more informative summary.
Common biostatistical scenarios and the preferred measure of central tendency
Data ScenarioRecommended MeasureBiostatistical Example
Nominal variableModeMost common blood type in a donor registry (e.g., O+)
Ordinal (e.g., pain scale 0–10)MedianMedian self-reported pain score after surgery
Symmetric, interval/ratioMean (± SD)Systolic blood pressure in a normotensive population
Right-skewed, interval/ratioMedian (IQR)Hospital length of stay, healthcare expenditures, viral load
Censored survival dataMedian survival timeMedian overall survival in an oncology trial (via Kaplan–Meier)

Worked Example — Serum Creatinine in a Renal Study

A nephrology research team measured serum creatinine (mg/dL) in 11 patients presenting with early-stage chronic kidney disease. The observed values were: 0.9, 1.1, 1.0, 1.3, 1.2, 1.1, 4.8, 1.0, 1.2, 1.1, 1.3. Note that one patient (4.8 mg/dL) has a markedly elevated value, suggesting advanced renal impairment — an outlier that will differentially affect the three measures.

Computing Mean, Median, and Mode for Serum Creatinine Data
1
Step 1 — Sort the Data in Ascending OrderArrange the 11 observations from smallest to largest: 0.9, 1.0, 1.0, 1.1, 1.1, 1.1, 1.2, 1.2, 1.3, 1.3, 4.8. Sorting is essential for identifying both the median position and the frequency of repeated values.
Sorted array: 0.9, 1.0, 1.0, 1.1, 1.1, 1.1, 1.2, 1.2, 1.3, 1.3, 4.8
2
Step 2 — Compute the MeanSum all values: 0.9 + 1.0 + 1.0 + 1.1 + 1.1 + 1.1 + 1.2 + 1.2 + 1.3 + 1.3 + 4.8 = 16.0. Divide by n = 11: x̄ = 16.0 / 11 ≈ 1.45 mg/dL. Notice this value is higher than most of the observations because the outlier (4.8) pulls the mean upward.
Mean = 1.45 mg/dL
3
Step 3 — Identify the MedianWith n = 11 (odd), the median is at position (11 + 1)/2 = 6th observation. Counting through the sorted array: the 6th value is 1.1. The median is unaffected by the outlier; replacing 4.8 with any value ≥ 1.1 would leave the median unchanged.
Median = 1.1 mg/dL
4
Step 4 — Determine the ModeCount frequencies: 0.9 appears 1×, 1.0 appears 2×, 1.1 appears 3×, 1.2 appears 2×, 1.3 appears 2×, 4.8 appears 1×. The value with the highest frequency is 1.1, so the dataset is unimodal.
Mode = 1.1 mg/dL
5
Step 5 — Interpret the ResultsThe mean (1.45) exceeds both the median and mode (1.1), signaling right skewness driven by the outlier. For this dataset, reporting the median (1.1 mg/dL) with the interquartile range more accurately represents the 'typical' patient than the mean. In a clinical context, the investigator might report: 'Median serum creatinine was 1.1 mg/dL (IQR: 1.0–1.3), consistent with Stage 2 CKD, although one patient had a markedly elevated value of 4.8 mg/dL, suggesting more advanced disease.'
Mean > Median = Mode → Right-skewed distribution → Prefer median for summary

Strengths, Limitations & Comparisons

No single measure of central tendency is universally superior; each carries a distinct set of strengths and weaknesses. The table below provides a systematic comparison along dimensions that matter most in biostatistical practice: sensitivity to outliers, algebraic tractability, applicability across measurement scales, and behavior under different distributional assumptions.

Comparative properties of the three classical measures of central tendency
PropertyMeanMedianMode
Uses all data pointsYes — every value contributesNo — depends only on rank orderNo — depends only on frequency
Sensitivity to outliersHighLowNone
Algebraic propertiesAdditive; unbiased estimator of μ; used in variance, ANOVANot additive; no closed-form in many distributionsNot algebraically tractable; may be non-unique
Measurement scalesInterval and ratio onlyOrdinal, interval, ratioAll scales including nominal
Sampling variabilityLowest (under normality)Slightly higher under normality; lower under heavy tailsHighly variable in small samples
Best suited when…Data are symmetric with no extreme outliersData are skewed, censored, or contain outliersData are categorical or you seek the most typical category
KEY TAKEAWAY
In biostatistical research, the mean dominates the analytical pipeline because it feeds directly into parametric tests (t-tests, ANOVA, regression). However, reporting the mean for skewed data without the median is like reporting only the average income in a neighborhood where a billionaire lives — technically correct but profoundly misleading. Best practice in journals like The Lancet and JAMA is to report the mean ± SD for symmetric data and the median with the interquartile range (IQR) for skewed data, clearly justifying the choice.

Connections to Advanced Statistical Theory

The mean, median, and mode are not isolated descriptive tools; they serve as conceptual and mathematical foundations for much of inferential biostatistics. Understanding how each measure connects to more advanced methods prepares you for the deeper analytical work encountered in clinical trial design, survival analysis, and machine learning for health data.

From descriptive measures to advanced inferential methods
Descriptive FoundationAdvanced ExtensionBiostatistical Application
Sample mean (x̄)Least-squares estimation; OLS regression coefficients minimize Σ(yᵢ − ŷᵢ)², extending the mean's squared-deviation minimization propertyModeling dose–response relationships, adjusting for confounders in RCTs
Sample medianQuantile regression; median regression minimizes Σ|yᵢ − ŷᵢ|, providing a robust alternative when residuals are non-GaussianModeling healthcare costs, length of stay, and other right-skewed outcomes
Median survival timeKaplan–Meier estimator; the median is the time at which the survival function S(t) = 0.50Primary endpoint in Phase III oncology trials; FDA approval decisions
Mode of a densityMaximum a posteriori (MAP) estimation in Bayesian inference; kernel density estimationIdentifying subpopulations in flow cytometry, pharmacokinetic modeling
Weighted meanMeta-analysis (inverse-variance weighting); generalized estimating equations (GEEs)Combining treatment effects across multi-site clinical trials

One especially important bridge concept is the Central Limit Theorem (CLT), which states that the sampling distribution of the mean approaches normality as sample size increases, regardless of the underlying population distribution (provided finite variance). This theorem justifies the widespread use of the mean in parametric inference. For the median, an analogous result holds: its sampling distribution is asymptotically normal with variance 1/(4nf(x̃)²), where f(x̃) is the population density evaluated at the true median. The practical implication is that both confidence intervals and hypothesis tests can be constructed around either measure for sufficiently large samples — though the choice of measure still affects interpretive meaning and statistical power.

Practice Problems

PROBLEM 1CONCEPTUAL
A public health researcher reports that the mean household income in a county is $78,000, while the median is $52,000. What does the large gap between these two values indicate about the income distribution, and which measure would be more appropriate to describe the 'typical' household income? Explain your reasoning.
PROBLEM 2BASIC CALCULATION
The following hemoglobin values (g/dL) were measured in 8 patients: 12.1, 13.4, 11.8, 14.0, 12.5, 13.4, 12.9, 13.4. Compute the mean, median, and mode.
PROBLEM 3INTERMEDIATE
A clinical researcher records the number of adverse events per patient in a Phase II trial: 0, 0, 1, 0, 2, 0, 1, 0, 0, 5. She wants to summarize this data for a regulatory submission. Calculate the mean, median, and mode, then recommend which measure to emphasize in the report and justify your recommendation.
PROBLEM 4APPLIED
An epidemiologist studying COVID-19 hospital stays in a metropolitan area collects the following lengths of stay (days) for 12 patients: 3, 4, 5, 5, 6, 7, 7, 8, 10, 14, 21, 35. She plans to present findings at a public health conference. (a) Compute all three measures of central tendency. (b) If she had to choose a single number for a policy brief aimed at hospital capacity planning, which measure should she use and why? (c) If one patient's 35-day stay is reclassified as a transfer (removed from the dataset), how does each measure change?
PROBLEM 5CRITICAL THINKING
A biostatistician analyzes time-to-event data from a cancer immunotherapy trial. Of 200 enrolled patients, 80 have died during follow-up and 120 are still alive (right-censored). The Kaplan–Meier curve has not yet crossed the 50% survival mark. The mean survival time cannot be computed without parametric assumptions. Explain (a) why the mean is problematic here, (b) whether the median can be reported, and (c) what alternative summary the team might report instead.

Summary — Mean, Median & Mode in Biostatistics

The three classical measures of central tendency — mean, median, and mode — each answer the question 'Where is the center?' through a different lens. The arithmetic mean (x̄ = Σxᵢ/n) is the balance point that minimizes squared deviations and feeds directly into parametric methods like t-tests, ANOVA, and regression, but is sensitive to outliers and skewness. The median is the positional middle value — robust to extreme observations and the preferred summary for skewed biostatistical data such as hospital costs, survival times, and viral loads. The mode identifies the most frequent value and is the only measure applicable to nominal data (blood type, genotype).

In practice, the relationship among the three measures serves as a quick skewness diagnostic: when mean > median, the distribution is right-skewed; when mean < median, it is left-skewed; when all three coincide, the distribution is symmetric. Best practice in biostatistical reporting is to choose the measure that matches your data's measurement scale and distributional shape, report it alongside an appropriate measure of spread (SD for the mean, IQR for the median), and justify the choice explicitly — as required by CONSORT, STROBE, and other reporting guidelines.

Varsity Tutors • Biostatistics • Mean, Median & Mode — Compute and interpret mean, median, mode