Module 5

Numerical Data Summaries

Introduction

Numerical data summaries are essential tools in statistics, enabling researchers and practitioners to describe, interpret, and compare datasets. In health sciences, these summaries help in understanding patterns, variability, and central values in biomedical and epidemiological data. This module focuses on the fundamental concepts and techniques for summarizing quantitative data, with applications in real-life health scenarios.

Measures of Central Tendency

Definition and Importance

Measures of central tendency provide a single value that best represents the overall data, indicating the center of the distribution. The most common measures are mean, median, and mode.

• Mean (Arithmetic Mean): The sum of all values divided by the number of observations. Sensitive to extreme values (outliers).

• Median: The middle value when data is ordered. Robust to outliers and skewed data.

• Mode: The most frequently occurring value. Useful for categorical or repetitive data.

Formulas:

• Mean: $\bar{x} = \frac{\sum_{i=1}^{n} x_i}{n}$

• Median: Middle value in ordered data; if even number of observations, average the two middle values.

• Mode: Value with the highest frequency.

Example: Daily calorie intake of 5 patients: 1800, 2000, 2200, 2100, 1900 kcal. Mean = $\frac{1800 + 2000 + 2200 + 2100 + 1900}{5} = 2000$ kcal.

Measures of Dispersion

Definition and Importance

Measures of dispersion describe the spread or variability of a dataset, indicating how much the data points differ from each other and from the central value.

• Range: Difference between the maximum and minimum values. $\text{Range} = \text{Max} - \text{Min}$

• Variance: Average squared deviation from the mean. Population variance: $\sigma^2 = \frac{\sum (x_i - \mu)^2}{N}$ Sample variance: $s^2 = \frac{\sum (x_i - \bar{x})^2}{n-1}$

• Standard Deviation: Square root of variance. Population: $\sigma = \sqrt{\sigma^2}$ Sample: $s = \sqrt{s^2}$

• Coefficient of Variation (CV): Standard deviation as a percentage of the mean. $\text{CV} = \frac{s}{\bar{x}} \times 100\%$

• Interquartile Range (IQR): Spread of the middle 50% of data. $\text{IQR} = Q_3 - Q_1$

Example: If the highest score is 95 and the lowest is 65, then the range is 30.

Shape of Distribution

Skewness (Asymmetry)

Skewness measures the asymmetry of a distribution. It indicates whether data are spread more to one side of the mean.

• Symmetric Distribution: Skewness = 0

• Positively Skewed (Right-Skewed): Skewness > 0

• Negatively Skewed (Left-Skewed): Skewness < 0

Formula (Sample Skewness): $\text{Skewness} = \frac{n}{(n-1)(n-2)} \sum \left( \frac{x_i - \bar{x}}{s} \right)^3$

Kurtosis (Peakedness)

Kurtosis measures the tailedness of a distribution, indicating the presence of outliers or extreme values.

• Mesokurtic: Normal kurtosis (Kurtosis ≈ 3)

• Leptokurtic: High kurtosis (fat tails, more outliers)

• Platykurtic: Low kurtosis (thin tails, fewer outliers)

Formula (Excess Kurtosis): $\text{Excess Kurtosis} = \text{Kurtosis} - 3$

Location Measures: Quantiles

Definition and Types

Quantiles divide a dataset into equal-sized parts, indicating the relative standing of values.

• Quartiles: Divide data into 4 parts (Q1, Q2, Q3)

• Deciles: Divide data into 10 parts (D1, D2, ..., D9)

• Percentiles: Divide data into 100 parts (P1, P2, ..., P99)

Example: In ordered data, Q1 is the value below which 25% of observations fall, Q2 is the median, and Q3 is the value below which 75% fall.

Grouped Data Summaries

Mean, Median, and Mode for Grouped Data

When data are grouped into classes, summary statistics are computed using class marks and frequencies.

• Mean: $\bar{x} = \frac{\sum f_i x_i}{\sum f_i}$, where $f_i$ is frequency and $x_i$ is class mark.

• Median: $\text{Median} = L + \left( \frac{\frac{n}{2} - F}{f} \right) w$, where L = lower boundary of median class, F = cumulative frequency before median class, f = frequency of median class, w = class width.

• Mode: $\text{Mode} = L + \left( \frac{f_1 - f_0}{2f_1 - f_0 - f_2} \right) w$, where $f_1$ = frequency of modal class, $f_0$ = frequency before modal class, $f_2$ = frequency after modal class.

Data Visualization: Box-and-Whisker Plot

Five-Number Summary

A boxplot visually summarizes data using five key statistics: minimum, Q1, median (Q2), Q3, and maximum. It highlights the spread, central tendency, skewness, and outliers.

• Interquartile Range (IQR): $\text{IQR} = Q_3 - Q_1$

• Fences for Outliers: Left fence: $L = Q_1 - 1.5 \times \text{IQR}$; Right fence: $R = Q_3 + 1.5 \times \text{IQR}$

Interpretation: The box shows the middle 50% of data. Whiskers extend to the lowest and highest values within the fences. Points outside the fences are considered outliers.

Comparing Distributions

Key Questions

• Which group has the higher median or mean?

• How do the spreads (IQR, range) compare?

• Are the distributions symmetric or skewed?

• Are there outliers, and how many?

• Do the groups overlap significantly?

Usefulness and Limitations of Summary Measures

Central Tendency

• Usefulness: Widely used, easy to interpret, good for comparing groups.

• Limitations: Sensitive to outliers, may not represent skewed data well.

Dispersion

• Usefulness: Indicates consistency, useful for quality control and risk assessment.

• Limitations: Some measures (e.g., range) are highly sensitive to outliers.

Quantiles

• Usefulness: Useful for growth charts, reference ranges, and diagnostics.

• Limitations: Not precise for small samples, sensitive to data grouping.

Applications in Health Sciences

Numerical summaries are indispensable in public health and clinical research for summarizing large datasets, identifying patterns, detecting outliers, and communicating findings. Proper selection and interpretation of summary measures are crucial for drawing valid conclusions, especially when data are skewed or contain outliers.

Summary Table: Measures of Central Tendency

Measure	Definition	Best Used When
Mean	Sum of all values divided by count	Data is symmetrical
Median	Middle value in ordered data	Data is skewed or has outliers
Mode	Most frequent value	Categorical or repetitive data

Summary Table: Measures of Dispersion

Measure	Definition	Usefulness	Limitations
Range	Max - Min	Simple, quick idea of spread	Highly sensitive to outliers
Standard Deviation	Square root of variance	Widely used, interpretable	Assumes normality, affected by outliers
IQR	Q3 - Q1	Robust to outliers	Does not use all data points

Summary Table: Shape of Distribution

Measure	Definition	Interpretation
Skewness	Asymmetry of distribution	0 = symmetric, >0 = right-skewed, <0 = left-skewed
Kurtosis	Tailedness of distribution	>3 = leptokurtic, =3 = mesokurtic, <3 = platykurtic

Additional info: Some formulas and applications were expanded for clarity and completeness, including grouped data formulas and health science examples.