Measures Of Central Tendency

1. Core Concept Overview & Intuition

A measure of central tendency is a single numerical value that summarizes an entire data set by identifying its "centre" or typical value. When data are collected and organized into a frequency distribution, the raw list of observations is often too large to interpret directly. A measure of central tendency condenses that information into one representative number around which the observations cluster. The three standard measures are the mean, the median, and the mode.

The mean is the arithmetic average. It is obtained by adding all observed values and dividing by the number of observations. Intuitively, the mean is the value each observation would receive if the total of all values were shared equally among all observations. Because it uses every observation in its computation, the mean is sensitive to the magnitude of each value, including unusually large or small ones.

The median is the middle value. When the observations are arranged in order of magnitude, the median is the value that divides the ordered data into two equal halves: half of the observations lie at or below it and half lie at or above it. The median depends only on the position of values in the ordered sequence, not on their numerical magnitude, so it is unaffected by how far the extreme values lie from the centre.

The mode is the most frequent value. It is the observation, or class of observations, that occurs more times than any other. The mode identifies the most common outcome in the data and is the only one of the three measures that can describe qualitative (categorical) data, since it relies on counting occurrences rather than on arithmetic.

These three measures answer subtly different questions about the same data. The mean answers "what is the total shared equally?"; the median answers "what value splits the ordered data in half?"; and the mode answers "what value occurs most often?". For a perfectly symmetric distribution the three coincide, but in general they differ, and the choice among them depends on the nature of the data and the purpose of the summary.

2. Mathematical Formalization & Core Definitions

> Definition (Arithmetic Mean). The arithmetic mean of a set of observed values is the sum of the values divided by the number of values. For observations $x_1, x_2, \ldots, x_n$, the mean, denoted $\bar{x}$, is

$$
\bar{x} = \frac{x_1 + x_2 + \cdots + x_n}{n}.
$$

Here $x_i$ denotes the $i$-th observed value and $n$ denotes the number of observations.

When the data are presented as a frequency distribution, each distinct value $x_i$ is accompanied by a frequency $f_i$, the number of times that value occurs. The mean is then computed as a weighted average:

$$
\bar{x} = \frac{f_1 x_1 + f_2 x_2 + \cdots + f_k x_k}{f_1 + f_2 + \cdots + f_k} = \frac{\sum_{i=1}^{k} f_i x_i}{\sum_{i=1}^{k} f_i},
$$

where $k$ is the number of distinct values, $f_i$ is the frequency of the value $x_i$, and $\sum_{i=1}^{k} f_i = n$.

For a grouped frequency distribution, the classes are first replaced by their mid-values (class marks), and the same weighted formula is applied to those mid-values. The arithmetic mean of a grouped frequency distribution is therefore computed by the formula above after the classes have been replaced by their mid-values.

> Definition (Median). The median is the middle value of a data set arranged in order of magnitude. For $n$ ordered observations, the median is the $\left(\frac{n+1}{2}\right)$-th value when $n$ is odd, and the average of the $\frac{n}{2}$-th and $\left(\frac{n}{2}+1\right)$-th values when $n$ is even.

For a grouped frequency distribution, the median is located within the median class, the class whose cumulative frequency first reaches or exceeds $\frac{n}{2}$. It is computed by interpolation within that class:

$$
\text{Median} = L_1 + \left(\frac{\frac{n}{2} - \left(\sum f\right)1}{f\right) c,}}
$$

where $L_1$ is the lower class boundary of the median class, $n$ is the total number of items in the data (total frequency, $\sum f$), $\left(\sum f\right)1$ is the sum of the frequencies of all classes lower than the median class, $f$ is the frequency of the median class, and $c$ is the size of the median class interval.}

> Definition (Mode). The mode is the most frequent value in a data set, that is, the value or class that occurs more times than any other. For a grouped frequency distribution, the modal class is the class with the maximum frequency, and the mode is computed by interpolation within it:

$$
\text{Mode} = L + \left(\frac{f_m - f_1}{2f_m - f_1 - f_2}\right) c,
$$

> where $L$ is the lower class boundary of the modal class, $f_1$ is the frequency of the class preceding the modal class, $f_2$ is the frequency of the class following the modal class, $f_m$ is the frequency of the modal class, and $c$ is the size of the modal class interval.

A data set may have one mode (unimodal), two modes (bimodal), or more; if every value occurs equally often, no mode exists.