Measures of Central Tendency: Mean, Median, and Mode

Last updated on:

When you look at a collection of numbers, your first instinct is often to find some way to summarize it. What’s typical? What represents this data best? These are the questions that measures of central tendency help answer. Whether you’re analyzing food safety inspection scores, customer satisfaction ratings, or employee performance metrics, understanding the mean, median, and mode gives you powerful tools to make sense of your data.

Table of Contents

What are measures of central tendency?

A measure of central tendency is a single value that attempts to describe a dataset by identifying its central position. These measures help you understand where most values in your distribution fall and are also called measures of central location. The three primary measures are the mean, median, and mode, and each calculates the central point using a different method.

The mean: Your arithmetic average

The mean is what most people call the average. It’s calculated by adding all values in your dataset and dividing by the number of values. For example, if you have test scores of 85, 90, 78, 92, and 88, the mean would be (85+90+78+92+88)/5 = 86.6.

The mean includes every value in your dataset as part of its calculation. This is both its strength and weakness. Because it considers all data points, any change in values will affect the mean. This makes it mathematically useful for further statistical analysis.

When the mean works best

The mean is your go-to measure when your data follows a normal distribution without extreme values. In symmetrical distributions, the mean, median, and mode are all equal, making the mean the preferred choice since it includes all values in its calculation.

The mean’s major limitation

The mean is highly sensitive to outliers. Consider a factory with ten employees: eight earn between $12,000 and $18,000 annually, while two executives earn $90,000 and $95,000. The mean salary would be $30,700, which doesn’t accurately represent what most workers earn. Extreme values pull the mean away from the typical value, making it misleading in such situations.

The median: Finding the middle ground

The median is the middle value when your data is arranged in order. If you have an odd number of observations, it’s simply the middle number. With an even number of observations, you take the average of the two middle values.

For the dataset 54, 54, 55, 56, 57, 58, 60, the median is 56 because it’s the value in the middle position. If we remove one value to get 54, 54, 55, 56, 57, 58, the median becomes (55+56)/2 = 55.5.

Why the median matters

The median is less affected by outliers and skewed data than the mean. It divides your distribution exactly in half, with 50% of observations on either side. This makes it particularly valuable when your data contains extreme values that would distort the mean.

Going back to our salary example, the median would be around $15,000, which much better represents what most workers actually earn. The median remains stable even when extreme values are present, providing a more accurate picture of the typical value.

Limitations to consider

While the median is robust against outliers, it cannot be used with categorical nominal data that cannot be logically ordered. Additionally, unlike the mean, the median is not amenable to further mathematical calculations, which limits its use in some statistical tests.

The mode: Spotting the most common value

The mode is the value that appears most frequently in your dataset. In the dataset 54, 54, 54, 55, 56, 57, 57, 58, 60, the mode is 54 because it occurs three times, more than any other value.

A dataset can have no mode (all values occur once), one mode (unimodal), two modes (bimodal), or multiple modes (multimodal). The mode has a unique advantage: it can be used for both numerical and categorical data. This makes it the only measure of central tendency available for nominal data.

When to use the mode

The mode is most useful when analyzing categorical data where you want to know the most common category. For instance, if you’re tracking the most frequent food safety violation in your inspections, the mode tells you which violation type appears most often. It’s also helpful for identifying peaks in your data distribution.

The mode’s drawbacks

The mode is rarely used as a summary statistic for continuous data because you’re unlikely to have repeated exact values. When the most common value is far from the rest of the data, the mode doesn’t accurately represent central tendency. Additionally, it’s not algebraically defined, limiting its use in statistical analysis.

Choosing the right measure for your data

Selecting the appropriate measure depends on your data type and distribution shape. For nominal categorical data, you must use the mode since you cannot calculate a mean or median from categories like “vanilla,” “chocolate,” or “strawberry.”

For ordinal data, the median is typically preferred, though the mode can also work. When dealing with interval or ratio data, your choice depends on the distribution.

In symmetrical distributions, the mean is often the preferred measure because it includes all values and any change in the data will affect it. However, when your data is skewed or contains outliers, the median becomes the better choice.

Understanding skewness and outliers

Skewness refers to asymmetry in your data distribution. In a positively skewed distribution, the tail extends to the right, and the mean is typically pulled toward higher values. In negatively skewed distributions, the tail extends left, pulling the mean toward lower values.

The relationship between these measures in skewed data follows a pattern. In right-skewed data, the mean is largest, followed by the median, then the mode. In left-skewed data, this order reverses. The median remains the most resistant to skewness, staying closer to the center of where most data points cluster.

Dealing with outliers

Outliers are extreme values that differ significantly from other observations. They can dramatically impact the mean while leaving the median relatively unchanged. Consider replacing a retirement age of 60 with 81 in a dataset where other values range from 54 to 60. The median might shift only slightly, but the mean would increase noticeably.

Despite their influence, outliers shouldn’t automatically be removed. If they represent valid extreme values, they’re part of your data story. The median offers a better representation of central tendency in these cases, though the mean can still be appropriate if the rest of your data is normally distributed.

Practical application in research

In real-world research, understanding which measure to use is crucial for accurate data interpretation. Income data, for instance, is typically reported using the median rather than the mean because higher earners create a false representation of typical income when expressed as a mean.

When analyzing food safety data, you might encounter scenarios where inspection scores cluster around certain values with occasional very high or low scores. The median would give you a better sense of typical performance than the mean, which might be skewed by those extreme scores. Meanwhile, the mode could tell you the most common score range, providing additional insight.

The key is to let your data’s characteristics guide your choice. Examine your distribution’s shape, check for outliers, and consider what question you’re trying to answer. Sometimes reporting multiple measures provides the most complete picture, allowing your audience to understand both the typical value and how the data is distributed.

What do you think? How might choosing between mean and median change the story your data tells? When working with skewed data in your field, which measure would give stakeholders the most honest picture of your findings?

How useful was this post?

Click on a star to rate it!

Average rating 0 / 5. Vote count: 0

No votes so far! Be the first to rate this post.

We are sorry that this post was not useful for you!

Let us improve this post!

Tell us how we can improve this post?

References
  1. https://statistics.laerd.com/statistical-guides/measures-central-tendency-mean-mode-median.php
  2. https://www.abs.gov.au/statistics/understanding-statistics/statistical-terms-and-concepts/measures-central-tendency
  3. https://online.stat.psu.edu/stat200/lesson/2/2.2/2.2.4/2.2.4.1
  4. https://statisticsbyjim.com/glossary/mean-vs-median/
  5. https://pmc.ncbi.nlm.nih.gov/articles/PMC3157145/
PDF ๐Ÿ“„

Comments

Leave a Reply

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

Research Methodology

1 Selection of Research Problem

  1. Science and Characteristics of Scientific Knowledge
  2. Need for Scientific Methodology
  3. Identification of Research Problem
  4. Statement of the Problem and Objectives

2 Review of Literature

  1. Review of Literature: Sources and Classification
  2. Uses of Review of Literature
  3. Steps in Review of Literature
  4. Writing Review of Literature and Theoretical Orientation
  5. Citation
  6. Writing Bibliographical Details of a Reference

3 Concept and Variables, Formulation and Testing of Hypothesis

  1. Concept, Construct and Variables
  2. Types of Variables
  3. Hypothesis
  4. Types and Forms of Hypothesis
  5. Characteristics, Function and Testing of Hypothesis

4 Research Design

  1. Characteristics of Research Design
  2. Criteria of a Research Design
  3. Max-Min-Con Principle
  4. Classification of Research Design
  5. Experimental Research Design
  6. Descriptive Research Design

5 Descriptive and Survey Research Design

  1. Characteristics of Descriptive Research Design
  2. Steps in Descriptive Research
  3. Aims of Descriptive Research Design
  4. Types of Descriptive Research Design
  5. Case Studies
  6. Observational Studies
  7. Historical Studies
  8. Field Studies
  9. Diagnostic Studies
  10. Explorative Studies
  11. Longitudinal Studies
  12. Correlational Studies
  13. Cross-Sectional Studies
  14. Action Research
  15. Evaluation Research
  16. Survey Research

6 Experimental Research

  1. Testing of hypothesis
  2. t-test
  3. ฯ‡2-test
  4. F-test
  5. Principles of Experimental Designs
  6. Completely Randomised Designs
  7. Randomized Complete Block Design
  8. Latin Square Design
  9. Factorial Experiments
  10. 2n factorial experiment
  11. 3n factorial experiment

7 Levels of Measurement

  1. Concept of Measurement
  2. Postulates of Measurement
  3. Nominal Scale
  4. Ordinal Scale
  5. Interval Scale
  6. Ratio Scale

8 Knowledge Test Constructions

  1. Knowledge Test
  2. Characteristics of a Good Test
  3. Steps in Standardised Test Construction
  4. Item Analysis
  5. Writing Test Items
  6. Preliminary Administration
  7. Reliability of the Final Test
  8. Validity of the Final Test
  9. Norms of the Final Test
  10. Item Difficulty and Discrimination

9 Data Collection

  1. Secondary Data Sources
  2. Instruments Used for Collecting Primary Data
  3. Validity, Data Editing, and Coding
  4. Data Tabulation and Presentation

10 Sampling Technique

  1. Importance of Sampling
  2. Types of Sampling Techniques
  3. Probability based Sampling Techniques
  4. Non-Probability based Sampling Techniques
  5. Sample Size Determination
  6. Sampling and Non-Sampling Errors

11 Quantitative Techniques

  1. Frequency Distribution
  2. Measures of Central Tendency
  3. Measures of Dispersion
  4. Correlation
  5. Regression
  6. Multiple Regressions
  7. Dummy Variable Analysis
  8. Discriminant Function Analysis
  9. Factor Analysis
  10. Principal Component Analysis

12 Qualitative Techniques

  1. Observation Method
  2. Interview Method
  3. Questionnaire Method
  4. Case Study Method
  5. Projective Techniques

13 Statistical Analysis and Packages

  1. ฯ‡2- test
  2. t-test
  3. F-test
  4. Basic Experimental Designs
  5. Factorial Experiments
  6. Non-Parametric Tests
  7. Run Test
  8. Sign Test
  9. Wilcoxon Signed Rank Test
  10. Mann-Whitney U-Test
  11. Kruskal-Wallis One-way Analysis of Variance
  12. Friedman Two-way Analysis of Variance

14 Report Writing

  1. Research Report
  2. Steps in Preparing the Report: Preliminary Considerations
  3. Main Components of a Research Report
  4. Diagrammatic Presentation
  5. Common Weaknesses in Research Report Writing