Mathematics
·
Why Your 'Average' Salary Is a Lie
Easy
5 min
Why Your 'Average' Salary Is a Lie
How the number that summarizes data can mislead you about salary statistics.
Share
Ten people walk into a bar. Nine of them earn $30,000 a year. One earns $10 million. The 'average' salary in that bar is now over $1 million. Does that feel right to you?
That bar scenario isn't a trick question — it's exactly how salary statistics get weaponized every day. A single billionaire in the dataset drags the average so far up it becomes meaningless. And yet 'average' is the number everyone reaches for by default. The real question isn't how to calculate it. It's when to trust it and when it's lying to your face.
Here's the setup. You have a pile of numbers and you need one number to represent them all. Three candidates show up for the job: mean, median, and mode. Each one tells a different story about the same data. Picking the wrong one doesn't just give you a bad answer — it gives you a confidently wrong answer, which is worse.
Analogy
Think about splitting a restaurant bill with friends. The mean is what everyone pays if you divide the total equally — fine when everyone ordered roughly the same thing. But if one person ordered lobster and three bottles of wine, suddenly the 'average' bill feels outrageous to the person who had a salad. The median is what the person in the middle of the group spent — a much fairer picture of what a 'typical' person at that table actually paid. The mode? That's the most common order. If six out of eight people got the burger, the burger is the mode. Useful if you're the restaurant planning tomorrow's menu. Less useful if you're splitting the bill.
Make your call
Back to that bar: nine people earning $30,000, one earning $10 million. What does the 'average' salary in that room actually tell you about what a typical person there earns?
A
It's a fair summary — the mean accounts for every person's income equally
B
It's misleading — the one high earner drags the average far above what almost everyone makes
C
It depends on the context — sometimes $1 million is a reasonable summary, sometimes not
So here's how each one actually works. The mean: add everything up, divide by how many things there are. The median: sort the numbers from smallest to largest, pick the one in the middle. If there's an even number of values, average the two middle ones. The mode: find whichever value shows up most often. That's it. The math is simple. The hard part — the part that actually matters — is knowing which one to use.
Key Insight
The mean gets wrecked by outliers. One extreme value — a billionaire, a freak storm, a viral post — drags it away from what's 'typical.' When your data has outliers, the median is almost always the more honest summary.
Example
Real example: U.S. household income. The mean household income sounds impressive — around $105,000. The median? About $74,000. That $31,000 gap exists entirely because a small number of ultra-wealthy households pull the mean upward. If a politician wants to make the economy sound great, they cite the mean. If a journalist wants to show how most people actually live, they cite the median. Same dataset. Completely different story — just like that bar full of people whose 'average' salary topped $1 million while nine of them took home $30,000. Now you know which question to ask when someone throws a statistic at you: 'Is that the mean or the median?'
Mode gets less attention, but it's the only one that works on non-numerical data. You can't calculate the mean of your camera roll by location — 'average city' isn't a thing. But you can find the mode: wherever you take the most photos. Retailers use mode constantly. They don't care about your average shoe size across all customers — they want to know which size to stock the most of. That's mode doing exactly what it's built for. A dataset can have no mode (all values appear once), one mode, or multiple modes. It's the messiest of the three, but sometimes it's the only one that makes sense.
Related threads
Mathematics
·
Why Half of Something Isn't Always Equal
Why Half of Something Isn't Always Equal
Sources
Images from Wikimedia Commons. Attribution required by license.
data charts numbers
Generated by Grok Imagine (xAI), prompt: Generate a single 16:9 horizontal image for an educational lesson.
Lesson title: "Why Your 'Average' Salary Is a Lie" · How the number that summarizes data can mislead you about salary statistics.
Dom…
Identifier: choiceemblemsnat00wynne (find matches)
Title: Choice emblems, natural, historical, fabulous, moral, and divine : for the improvement and pastime of youth : displaying the beauties and morals of the ancient fabulists, the whole calculated to convey the golden lessons of instruction under a new and more delightful dress : written for the amusement of a young nobleman
Year: 1793 (1790s)
Authors: Wynne, John Huddlestone, 1743-1788
Subjects: Conduct of life --Juvenile poetry. Conduct of life --Juvenile literature. Emblem books, English --Juvenile literature.
Publisher: London : Printed by J. Chapman for E. Newbery, corner of St. Paul's Church-yard
Contributing Library: University of Illinois Urbana-Champaign
Digitizing Sponsor: University of Illinois Urbana-Champaign
View Book Page: Book Viewer
About This Book: Catalog Entry
View All Images: All Images From Book
Click here to view book online to see this illustration in context in a browseable online version of this book.
Text Appearing Before Image:
found except in the Paths of Virtue^if you deviate from Her ways, you muftnever expert to find the Road to Happi-nefs:—you will become a Wanderer, andthe hope of your Piigiimagc will be loft. For thefe three are as inieparable as, fire,light, and heat; where the one i?, thereyou will find the others ; and the rewardIhall be fuch as will far tranfcend the painsyou may be at in acquiiing fuch an inefti-raabie Treafure. In Ulfl- ( ^86 ) In the mean time, envy not the acqttions of others; for that is bafe and feifiih*;neither fay within yourfelf, Such an one ishappy ^ nxihilfl I am expo fed to adverjify. For you know not the fecrets of menshearts; and it may be, that the perfonwhom you efteem happy^ is a prey to corrod-ing grief-i and pines in fecret anxiety. Atleaft, know this, That the ftatc of no.human being can be determined, till Deathcloff s the fcene ;—and the lall end of theGoodo^Xy can be Happy, Emulate theirvirtues^ and, doubtlcfs, you fiiall fliarc intheir felicfiy* EMBLEM
Text Appearing After Image:
EMBLEM LXIII. Of the U/e of Time, ^RUE to the Sun the Dial ftil! abides, And points Times coufe minutely as it glides;Ths bids us haflen to be wife, and fhowHow rapid in their courfe the minutes uow;Seize on the winded hours without delay,Njr trufl to-morrow v/hile you Jive to-day. MORAL. Time well em^loyd is a moft- certain giif^jEaincil: of pleafure, remedy for pain;The chief of biefljnjrs on its courfe attends,Sicce on its uFe Eternity depends. BEHOLD ( i8S ) BEHOLD how true the Dial is to theSun, and how exa/lly it marks the hours,whofe courfe might otherwife pafs unno-ticed or unknown. This ufeful invention we owe to the ma-thematicians, of ancient days, who thus fur-niflied men with the means of accuratelydiftinguifhing the difierent parts of the day,and dividing them into equal portions,whereby labour and reft, ft.udy and amufe-ment, were better regulated, and the wafteof time feen in a moment, without the troubleof tedious calculations. The Romans (mafters of the world) wereat
Note About Images
Please note that these images are extracted from scanned page images that may have been digitally enhanced for readability - coloration and appearance of these illustrations may not perfectly resemble the original work.
Internet Archive Book Images · No restrictions
Quiz yourself