Outliers in Statistics: How to Find Them Fast (IQR Method)

Key Takeaways

• Outliers in statistics are values far from most of the data, but they are not always “wrong.”
• The 1.5 IQR rule is a common, quick way to flag possible outliers.
• Don’t delete outliers automatically. First, check for errors or special reasons.

Outliers in statistics can feel scary because one weird number can change your answer. You’re not behind. You just need a clear method.

You’ll learn the IQR method step-by-step and see worked examples you can copy for homework and tests.

Quick answer

Outliers in statistics are data values that sit far away from the rest. A fast way to find them is the 1.5 IQR rule:
Step 1: Find Q1 and Q3 (the 25th and 75th percentiles).
Step 2: Compute IQR = Q3 − Q1.
Step 3: Compute fences: Lower = Q1 − 1.5·IQR, Upper = Q3 + 1.5·IQR.
Values outside the fences are possible outliers.

What is an outlier?

Outlier (a value far from most values) is a number that looks unusually low or high compared to the rest of the data.

Anomaly vs extreme value

• Anomaly (a value that may be an error): like a typo, wrong unit, or broken measurement tool.
• Extreme value (a real but rare value): unusual, but still true (like a very tall person in a class).

Important: The IQR rule flags values. It does not prove they are “bad.”

The IQR method (1.5 IQR rule)

IQR (interquartile range): the spread of the “middle half” of your data (Q3 − Q1).

Formula table

This is the rule used in many classes and box plots.

Step-by-step: How to find outliers

Steps

1. Sort the data from least to greatest.
2. Find Q1 and Q3. (Use your class rule for quartiles.)
3. Compute IQR = Q3 − Q1.
4. Compute the fences:
   • Lower fence = Q1 − 1.5·IQR
   • Upper fence = Q3 + 1.5·IQR
5. Compare each value to the fences.
   • If value < lower fence → possible low outlier
   • If value > upper fence → possible high outlier
6. Quick self-check:
   • Are Q1 < median < Q3?
   • Is IQR positive? (It should be.)

What to write on your paper (mini-checklist)

• Sorted data: ________
• Q1 = ___, Q3 = ___
• IQR = Q3 − Q1 = ___
• Lower fence = Q1 − 1.5(IQR) = ___
• Upper fence = Q3 + 1.5(IQR) = ___
• Outliers (if any): ________

Worked examples

Example 1

Data: 2, 3, 3, 4, 4, 5, 20

1. Sorted: 2, 3, 3, 4, 4, 5, 20
2. Q1 = 3, Q3 = 5 (middle halves)
3. IQR = 5 − 3 = 2
4. Lower fence = 3 − 1.5(2) = 3 − 3 = 0
   Upper fence = 5 + 1.5(2) = 5 + 3 = 8
5. Compare: 20 > 8 → 20 is a possible outlier

Why it happens: 20 is far above the main group (2 to 5).

Example 2

Quiz scores: 61, 68, 70, 72, 74, 75, 77, 78, 79, 95

1. Sorted: 61, 68, 70, 72, 74, 75, 77, 78, 79, 95
2. Lower half (first 5): 61, 68, 70, 72, 74 → Q1 = 70
   Upper half (last 5): 75, 77, 78, 79, 95 → Q3 = 78
3. IQR = 78 − 70 = 8
4. Lower fence = 70 − 1.5(8) = 70 − 12 = 58
   Upper fence = 78 + 1.5(8) = 78 + 12 = 90
5. Compare: 95 > 90 → 95 is a possible outlier

One-line meaning: That 95 is unusually high compared to most scores.

Example 3 (exam-style “trap”: skewed data)

Minutes spent studying: 5, 7, 8, 9, 10, 12, 15, 18, 40

1. Sorted: 5, 7, 8, 9, 10, 12, 15, 18, 40
2. Q1 = 7.5 (median of 5,7,8,9) → (7 + 8)/2
   Q3 = 16.5 (median of 12,15,18,40) → (15 + 18)/2
3. IQR = 16.5 − 7.5 = 9
4. Lower fence = 7.5 − 1.5(9) = 7.5 − 13.5 = −6
   Upper fence = 16.5 + 1.5(9) = 16.5 + 13.5 = 30
5. Compare: 40 > 30 → 40 is a possible outlier

Trap tip: Even if the data is “right-skewed” (long tail), you still apply the same fence rule. Then you interpret carefully.

Sale

TUL® Mechanical Pencils, 0.7 mm, Black Barrels,...

$12.94

Buy on Amazon

Pentel Twist Erase III Mechanical Pencils 0.7mm -...

$15.79

Buy on Amazon

Sale

BIC Xtra Strong Mechanical Pencils with Erasers,...

$6.08

Buy on Amazon

Practice set (with answer key)

Directions: Use the 1.5 IQR rule to find possible outliers.

1. 1, 2, 2, 3, 3, 4, 4, 12
2. 10, 11, 12, 12, 13, 13, 14, 15, 16
3. 5, 6, 6, 7, 8, 9, 10, 30
4. 42, 44, 45, 46, 46, 47, 48, 49, 50, 90
5. Trap: 2, 2, 2, 2, 2, 2, 2, 9
6. 100, 102, 103, 104, 105, 106, 107, 108, 140
7. 3, 4, 4, 4, 5, 5, 6, 7, 8
8. 12, 13, 13, 14, 15, 16, 16, 17, 60

Answer key (outliers only)

1. 12
2. none
3. 30
4. 90
5. 9 (high outlier)
6. 140
7. none
8. 60

Short solutions (2 items):

• #1: Q1 = 2, Q3 = 4 → IQR = 2 → upper fence = 4 + 3 = 7 → 12 > 7 → outlier.
• #5: Q1 = 2, Q3 = 2 → IQR = 0 → fences both = 2 → 9 > 2 → flagged (this is why “IQR = 0” sets a very strict fence).

Common mistakes

If you get stuck…

• “I don’t know what formula to use.”
  Use the fence formulas: Q1 − 1.5·IQR and Q3 + 1.5·IQR.
• “My IQR is 0.”
  That means Q1 = Q3 (many repeated values). The fences may become very tight, so one different value gets flagged.
• “My quartiles don’t match my friend’s.”
  Some classes use slightly different quartile rules. Use the rule your teacher taught (or your textbook’s rule).
• “My calculator gives a different value.”
  Tools may use a different quartile method. If it’s graded, follow your class method.
• “I’m not sure if I should remove the outlier.”
  Don’t remove it just because it’s flagged. Check for data entry or measurement errors first. If it might be real, keep it and explain it.
• When to get help:
  If your dataset is from a research project and removing a value changes your conclusion, ask your teacher/adviser what rule to follow.

Should you remove outliers?

Often, no—not right away.

A good, simple decision path:

1. Could it be an error? (typo, wrong unit, broken tool)
   • If yes, fix it if you can. If you can’t, you may remove it with a note.
2. Could it be real and meaningful?
   • If yes, keep it. It may be important, not “bad.”
3. If you’re doing research:
   • Don’t delete points just because a rule flagged them. You need a reason you can explain.

Next steps

• Descriptive Statistics Calculator — Use it to compute quartiles and IQR faster when you’re checking homework.
• Box plot interpretation — Learn how outliers show up as dots beyond the whiskers.
• Skewness — Helps you explain why some data naturally has a long tail (so “extremes” happen more).
• Range vs interquartile range — Learn when range is misleading and why IQR is often more stable.

How we know

• The 1.5·IQR fence rule is a standard classroom method for flagging possible outliers.
• The “possible outlier” idea matters: flagged values should be investigated, not blindly deleted.
• Steps and examples are aligned with widely used intro statistics teaching materials and practice sets.

Use this the right way

Use this guide to learn the method and show your work. Don’t copy answers without understanding. If this is for a research project, keep notes on any data you change and cite your sources.

Study tools that can help

• If you get distracted easily: try a simple timer (25 minutes study, 5 minutes break).
• If you lose steps on paper: use a dedicated math notebook so your quartile work stays organized: High-Quality Notebooks for Math Notes
• If you want faster checking: use the Descriptive Statistics Calculator to verify quartiles and IQR after you do it by hand.

References

Khan Academy. (n.d.). Identifying outliers with the 1.5xIQR rule. https://www.khanacademy.org/math/statistics-probability/summarizing-quantitative-data/box-whisker-plots/a/identifying-outliers-iqr-rule

OpenStax. (2023). Introductory statistics 2e: Measures of the location of the data (IQR and potential outliers). https://openstax.org/books/introductory-statistics-2e/pages/2-3-measures-of-the-location-of-the-data