Fundamentals of Social Statistics by Adam J. McKee

The mean (symbolized  “X -bar” and sometimes M as a sample statistic, or μ as a population parameter) is the average that is the “balance point” in a distribution.  It is calculated by adding up (summing) all of the scores and dividing by the number of scores (n).  The mean provides researchers with a way of finding the most typical value in a set of scores.

The mean is like a compass in the vast sea of numbers. It helps point us to a central value, offering a quick snapshot of where the data lies. For social science students, it’s one of the first tools in your statistical toolkit.

Definition and Basics

Think of the mean as a type of average. Let’s say different police precincts reported varying numbers of arrests for a specific crime over a month. Instead of remembering the arrests from each precinct, you could just consider the mean number of arrests to get an overall sense of the situation. It’s determined by adding all reported arrests and then dividing by the number of precincts.

For example, if three precincts reported 10, 20, and 30 arrests, respectively, for a particular crime in a month, the mean would be (10+20+30) ÷ 3 = 20 arrests.

Computing the Mean

Here’s how you do it:

  1. Add up all the numbers.
  2. Divide by the number of items you have.

In statistical notation, the formula is as follows:

X̄ = (ΣX) / N


  • X̄ is used to signify the “mean of X”
  • Σ is the summation symbol (“add up all of the X scores”).
  • X represents the individual data points (scores).
  • N is the total number of data points.

Properties of the Mean

Sensitivity to Extreme Values

One notable characteristic of the mean is its likeness to a seesaw, especially when it comes to handling outliers. In a dataset, if you introduce a particularly high or low value, the mean tends to shift in response, much like how a seesaw would tilt under the weight of a heavier individual. Such extreme values are termed “outliers.”

Consider our previous example: if we added a fourth score of 100, the mean would dramatically shift from 20 to 40, showcasing how a single outlier can influence the overall average.

Representative of Total Value

Another distinct feature of the mean is that it considers every value in the dataset. This ensures that the mean is representative of the total sum of values. For instance, if you have a set of salaries, the mean salary will always multiply back to the total combined salary of all individuals when multiplied by the number of individuals.

Affected by Sample Size

The size of your dataset or sample can also influence the mean. If you have a small sample, an outlier can have a more pronounced effect on the mean compared to when it’s a part of a larger dataset. Thus, the context in which the mean is calculated is crucial.


The mean has a linear property. If you were to add (or subtract) a constant value to each data point in your set, the mean would increase (or decrease) by the same constant. For example, if everyone in a class got a bonus 5 points on a test, the mean score would also rise by 5 points.

Advantages and Limitations

The mean, often colloquially known as the average, stands as one of the most straightforward and commonly employed statistical measures. Its simplicity stems from its calculation: summing all values and then dividing by the count of values. This straightforwardness makes it an easily comprehensible measure for those not deeply versed in statistics. Because of its widespread recognition and ease of understanding, it serves as the go-to choice for many researchers, analysts, and professionals across fields.

However, every tool, no matter how widely used, has its strengths and weaknesses, and the mean is no exception. One notable limitation of the mean is its sensitivity to extreme values, often termed outliers. To illustrate, consider a scenario where nine people have an income of $30,000, and one person has an income of $1,000,000. The mean income in this scenario would be significantly higher than the income of nine out of ten people in the group. In such cases, the mean might paint a distorted picture of the central tendency, potentially leading to misinterpretations.

In instances where data is skewed by extreme values, other measures of central tendency, such as the median, can provide a more accurate depiction. The median, which represents the middle value in a data set when all values are arranged in ascending or descending order, is resistant to the influence of outliers. Thus, when data doesn’t evenly spread around a central value or is significantly skewed, turning to the median or even the mode might be more insightful.

It’s worth noting, however, that the mean isn’t merely popular due to its simplicity. It holds significant utility in various advanced statistical procedures, particularly hypothesis testing—a fundamental concept in inferential statistics. Hypothesis tests often compare means from different samples or groups to draw conclusions about populations. We’ll delve deeper into this in later chapters. In essence, the key is to use the mean wisely. When the data meets the assumptions of the mean, such as being normally distributed without extreme outliers, the mean remains an invaluable and powerful tool in data analysis.


Pitfalls and Precautions

The mean, celebrated for its straightforward nature and ubiquity in statistics, provides a quick snapshot of the central value in a dataset. It’s a beacon that points to the ‘middle ground’ of our data, and for many, it offers a digestible summary of a collection of numbers. This quality, combined with its easy computation, makes the mean a staple in various fields—from social sciences to finance.

However, like any tool, its effectiveness is contingent upon its proper application. One of the most commonly encountered pitfalls with the mean is its susceptibility to extreme values. Outliers, or values that significantly deviate from the rest, have the potential to dramatically influence the mean. For instance, in a small town where most houses cost between $150,000 and $200,000, a single mansion priced at $2 million can drastically inflate the mean house price. This skewed mean might then misrepresent the general affordability of housing in that town.

Such scenarios underscore the importance of viewing the mean as one piece in the broader puzzle of data analysis. While it provides a general sense of where the data lies, it doesn’t always capture the entire story. In situations where outliers are present, or where data is notably skewed, the median (the middle value when data is sorted) or even the mode (the most frequently occurring value) might offer clearer insights. These alternative measures of central tendency can act as balancing weights, helping to provide a more rounded understanding of the dataset.

Now, returning to the mean, it’s imperative to understand its underlying assumptions. When we employ the mean to represent a dataset, we’re operating under several foundational premises. First and foremost, we assume the data operates on an interval or ratio scale of measurement. These scales, distinct from nominal and ordinal scales, permit meaningful calculation of an average since they represent equidistant measures and true zero points, respectively. Next, we anticipate that the data is roughly symmetrically distributed around the center, meaning there aren’t extreme outliers that can bias the mean. Furthermore, for the mean to be truly representative of a dataset, it’s optimal for the data to be normally distributed. These assumptions, when satisfied, ensure that the mean effectively embodies the central location of our data. If these assumptions are not met, it’s prudent to exercise caution when relying solely on the mean as the representative value.


The mean, often termed the “average,” is a fundamental concept in statistics. At its core, it provides a quick glimpse into the central value of a dataset. Calculating it involves summing all values in a set and dividing by the count of these values. Simple as it may sound, this tool is not without its nuances.

In the realm of statistics, the mean is akin to a flashlight guiding us through the dark expanse of data. For instance, in a crime dataset, if we want to ascertain the typical number of incidents in various precincts, the mean could point us to an overarching trend. However, its brilliance can sometimes be clouded by extreme values, or outliers. A single aberrant value, either exceptionally high or low, can tilt the mean, potentially distorting our perception of the data’s central value.

But why does this sensitivity to outliers matter? Consider a town’s average income. While most residents might earn a similar amount, a few multi-millionaires can significantly elevate the mean income, painting a potentially misleading picture of the town’s economic landscape. In such skewed scenarios, other measures like the median might offer a more accurate perspective.

Yet, despite its vulnerabilities, the mean remains indispensable, especially in advanced statistical procedures. It plays a pivotal role in hypothesis tests, a topic to be unraveled in subsequent chapters. Such tests frequently compare means from different groups, offering insights about broader populations.

However, for the mean to serve effectively, certain assumptions must be met. It’s ideal for the data to be on an interval or ratio scale, which ensures equal spacing between values and a genuine zero point. Moreover, a roughly symmetrical data distribution around the central value, without extreme outliers, is preferable. Adhering to these prerequisites ensures that the mean truly shines as an emblematic representation of our data.


[ Back | Contents | Next ]

Last Modified:  09/27/2023

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.