Mean, Median, Mode, Variance and SD: JEE Main Notes
Reviewed Last updated 2026-11-20. Syllabus topic: Mean, median, mode and measures of dispersion.
In short. Mean, median and mode locate the centre of data; mean deviation, variance and standard deviation measure its spread. For grouped data use class marks with frequencies. Variance = mean of squares minus square of the mean.
- Mean = sum of f x / sum of f; for grouped data x is the class mark, and a step-deviation method shortens the work.
- Median of grouped data = l + ((N/2 - cf)/f) h, and mode = l + ((f1 - f0)/(2 f1 - f0 - f2)) h.
- Variance = (sum of f x²)/N - (mean)²; the standard deviation is its positive square root, with the same unit as the data.
- Shifting every value by c leaves variance and SD unchanged; scaling by k multiplies the variance by k² and the SD by the modulus of k.
- Combined variance of two groups needs the group means: add n (variance + offset²) for each group, then divide by the total number.
Syllabus line: STATISTICS AND PROBABILITY : Measures of dispersion; calculation of mean, median, mode of grouped and ungrouped data, calculation of standard deviation, variance and mean deviation for grouped and ungrouped data. Probability: Probability of an event, addition and multiplication theorems of probability, Baye's theorem, probability distribution of a random variable. (NTA JEE (Main) 2026 syllabus, Mathematics, Unit 13 (https://jeemain.nta.nic.in/document/syllabus-2026/))
Before you start
Prerequisites
- Sequences and series: sums of natural numbers and their squares (lesson planned): Variance of the first n natural numbers uses the sum of squares. Read this first if those sum formulae are not ready.
- Sets, relations and functions: reading tables and linear maps (lesson planned): Change of origin and scale is a linear map y = a x + b. Read this first if linear maps on a set of values feel unfamiliar.
Related topics
- Statistics and Probability (lesson planned)
- Probability, addition and multiplication theorems
- Sequences and Series (lesson planned)
The concept step by step
Mean of ungrouped and grouped data
The arithmetic mean of n observations is the sum divided by n. If the value x_i occurs f_i times, the mean is (sum of f_i x_i)/(sum of f_i). For a class interval the value used is the class mark, the midpoint of the class. If the numbers are large, choose an assumed mean a, write d = x - a and use mean = a + (sum of f d)/(sum of f); the result does not depend on the choice of a.
Median and mode
The median is the middle value after ordering: the middle term for odd n, and the average of the two middle terms for even n. For a grouped table find N, locate the first class whose cumulative frequency reaches N/2 and apply median = l + ((N/2 - cf)/f) h, where l is its lower limit, cf the cumulative frequency before it, f its frequency and h the width. The mode is the most frequent value; for grouped data the modal class has the largest frequency and mode = l + ((f1 - f0)/(2 f1 - f0 - f2)) h.
Mean deviation
Mean deviation about a point a is the mean of the absolute deviations, (sum of f |x - a|)/(sum of f). It is least when a is the median, and the examiner may ask it about the mean or about the median, so read the question. Compute the central value first, write each |x - a| in a column, multiply by frequency and divide by the total frequency. The unit is the unit of the data.
Variance and standard deviation
Variance is the mean of the squared deviations from the mean: variance = (sum of f (x - mean)²)/N. The shortcut variance = (sum of f x²)/N - (mean)² avoids subtracting the mean in every row. The standard deviation is the positive square root of the variance and has the data unit. In JEE Main the divisor is N (the number of observations), not N - 1.
Effect of changing origin and scale
If y = a x + b then mean of y = a (mean of x) + b, variance of y = a² (variance of x) and SD of y = modulus of a times SD of x. A shift b does not change the spread at all. This is why a step-deviation u = (x - a)/h is allowed: compute the SD of u and multiply by h to return to the SD of x.
Combining groups and correcting data
For two groups with sizes n1, n2, means m1, m2 and variances v1, v2, first find the combined mean m. Then combined variance = (n1 (v1 + (m1 - m)²) + n2 (v2 + (m2 - m)²))/(n1 + n2). For a wrongly recorded value, correct the sum and the sum of squares separately, then recompute the mean and the variance; changing only the mean is a common error.
Worked examples
Each solution shows the marks every step would earn. The sanity check tests units and size of the answer.
Example 1 (easy)
Find the mean of the numbers 12, 15, 18, 21, 24 and 30.
| Step | Marks |
|---|---|
| Add the values: 12 + 15 + 18 + 21 + 24 + 30 = 120. | 1 |
| Divide by the number of values, 6. | 1 |
| Mean = 120/6 = 20. | 1 |
| Total: 3 marks | 3 |
Answer: 20
Fast method: Take 20 as an assumed mean: deviations -8, -5, -2, 1, 4, 10 add to 0, so the mean is exactly 20.
Sanity check: The mean lies between the smallest value 12 and the largest value 30, and the deviations from 20 sum to zero as they must.
Example 2 (easy)
Find the mode of grouped data with classes 10 to 20, 20 to 30, 30 to 40 and 40 to 50 with frequencies 5, 12, 18 and 9.
| Step | Marks |
|---|---|
| The modal class is 30 to 40, with the highest frequency 18, so l = 30, f1 = 18, f0 = 12, f2 = 9, h = 10. | 1 |
| Mode = l + ((f1 - f0)/(2 f1 - f0 - f2)) h = 30 + (6/15) x 10. | 1 |
| Mode = 30 + 4 = 34. | 1 |
| Total: 3 marks | 3 |
Answer: 34
Fast method: The mode sits above the lower limit by h times (f1 - f0)/(2 f1 - f0 - f2) = 10 x 6/15 = 4 units.
Sanity check: The mode 34 lies inside the modal class 30 to 40; it is nearer the lower end because the previous class (12) is more populated than the next one (9).
Example 3 (medium)
For the data 3, 5, 7, 8, 9, 12 find the mean deviation about the median.
| Step | Marks |
|---|---|
| The data is already ordered and n = 6, so the median = (7 + 8)/2 = 7.5. | 1 |
| Absolute deviations from 7.5 are 4.5, 2.5, 0.5, 0.5, 1.5, 4.5, which add to 14. | 2 |
| Mean deviation = 14/6 = 2.33. | 1 |
| Total: 4 marks | 4 |
Answer: 2.33
Fast method: Pair the values around the median: (3, 12) gives 4.5 + 4.5, (5, 9) gives 2.5 + 1.5 and (7, 8) gives 0.5 + 0.5, so the total is 14.
Sanity check: Units are those of the data. The mean deviation 2.33 is below the SD (about 2.87) as it always is, and no deviation exceeds the range 9.
Example 4 (medium)
Find the standard deviation of the data 2, 4, 4, 4, 5, 5, 7, 9.
| Step | Marks |
|---|---|
| Mean = 40/8 = 5. | 1 |
| Squared deviations: 9, 1, 1, 1, 0, 0, 4, 16, which add to 32. | 1 |
| Variance = 32/8 = 4. | 1 |
| SD = square root of 4 = 2. | 1 |
| Total: 4 marks | 4 |
Answer: 2
Fast method: Mean of squares = (4 + 16 + 16 + 16 + 25 + 25 + 49 + 81)/8 = 232/8 = 29, and 29 - 25 = 4.
Sanity check: SD has the data unit. The largest deviation is 4 and the SD 2 is of the order of the typical gap; a zero or negative variance would signal an arithmetic error.
Example 5 (medium)
Find the standard deviation of the grouped data with class marks 5, 15, 25, 35 and frequencies 4, 6, 8, 2.
| Step | Marks |
|---|---|
| N = 20. Sum of f x = 20 + 90 + 200 + 70 = 380, so the mean = 19. | 1 |
| Sum of f x² = 4(25) + 6(225) + 8(625) + 2(1225) = 100 + 1350 + 5000 + 2450 = 8900. | 1 |
| Variance = 8900/20 - 19² = 445 - 361 = 84. | 1 |
| SD = square root of 84 = 9.17. | 1 |
| Total: 4 marks | 4 |
Answer: about 9.17
Fast method: Step deviation u = (x - 25)/10 gives u = -2, -1, 0, 1 with f = 4, 6, 8, 2. Mean of u = -0.6 and mean of u² = 1.2, so variance of u = 1.2 - 0.36 = 0.84, SD(u) = 0.9165 and SD(x) = 10 x 0.9165 = 9.17.
Sanity check: The SD is about one third of the range of class marks (30), reasonable for data spread over four classes; the SD has the same unit as marks.
Example 6 (medium)
Find the variance of the first 10 natural numbers.
| Step | Marks |
|---|---|
| Mean = (10 + 1)/2 = 5.5. | 1 |
| Sum of squares = 10 x 11 x 21/6 = 385, so the mean of squares = 38.5. | 1 |
| Variance = 38.5 - 5.5² = 38.5 - 30.25 = 8.25. | 1 |
| Total: 3 marks | 3 |
Answer: 8.25
Fast method: Use the standard result (n² - 1)/12 = 99/12 = 8.25.
Sanity check: Limiting case n = 1 gives variance 0, as for a single observation; for n = 10 the SD is 2.87, plausible for values spread from 1 to 10.
Example 7 (hard, multi-step)
Twenty observations have mean 15 and standard deviation 3. It is found that the value 8 was recorded as 18. Find the correct variance.
| Step | Marks |
|---|---|
| Wrong sum = 20 x 15 = 300; wrong sum of squares = 20 x (9 + 225) = 4680. | 2 |
| Correct sum = 300 - 18 + 8 = 290, so the correct mean = 14.5. | 1 |
| Correct sum of squares = 4680 - 324 + 64 = 4420. | 2 |
| Correct variance = 4420/20 - 14.5² = 221 - 210.25 = 10.75. | 1 |
| Total: 6 marks | 6 |
Answer: 10.75
Fast method: Work with totals: sum and sum of squares of the 20 values, change the two totals, then variance = (sum of squares)/20 - (mean)².
Sanity check: The variance rises from 9 to 10.75 because the true value 8 lies farther from the mean (6.5 below 14.5) than the wrong value 18 did (3 above 15). Units: the variance is in the squared data unit.
Formulas and units
| Result | Formula | Valid when | Units or note | Worked example |
|---|---|---|---|---|
| Mean (grouped) | mean = (sum of f x)/N, N = sum of f; x = class mark | grouped data, x is the class mark, N is total frequency | same as data | Example 5 |
| Assumed mean method | mean = a + (sum of f d)/N, d = x - a | grouped or ungrouped data; a is any chosen assumed mean | same as data | none |
| Median (grouped) | l + ((N/2 - cf)/f) h | continuous classes; l lower limit of median class, cf before it | same as data | none |
| Mode (grouped) | l + ((f1 - f0)/(2 f1 - f0 - f2)) h | continuous classes; modal class has highest frequency, equal widths | same as data | Example 2 |
| Mean deviation about a | (sum of f |x - a|)/N | about any a; use median or mean as a; N is total frequency | same as data | Example 3 |
| Variance | (sum of f x²)/N - (mean)² | variance divides by N (not N - 1); the mean is of the same data | data unit squared | Example 5 |
| Standard deviation | square root of variance | variance is non-negative; SD is the positive root | same as data | Example 4 |
| Linear change y = a x + b | mean_y = a mean_x + b; var_y = a² var_x | a and b constants; SD scales by |a|, a shift does not change it | n/a (as data) | none |
| First n natural numbers | mean (n + 1)/2; variance (n² - 1)/12 | data 1, 2, ..., n, each once; variance divides by n | pure numbers | Example 6 |
| Combined variance | (n1 (v1 + (m1 - m)²) + n2 (v2 + (m2 - m)²))/(n1 + n2) | two groups with sizes n1, n2, means m1, m2, variances v1, v2; m is combined mean | data unit squared | none |
| Coefficient of variation | (SD/mean) x 100 | mean not zero, usually positive data | percent | none |
Common mistakes and speed
- Taking the class with the highest cumulative frequency, or the highest frequency, as the median class. Why it tempts: Both words look alike. The median class is the first class where the cumulative frequency reaches N/2; the highest-frequency class is the modal class.
- Using the cumulative frequency of the median class itself as cf in the median formula. Why it tempts: The table shows the cumulative frequency in the same row as the median class, so the nearest number is picked; the formula needs the cumulative frequency of the class before it.
- Dividing the sum of squared deviations by n - 1. Why it tempts: Some textbooks define the sample variance that way; the syllabus form divides by n, and the option with n - 1 is placed among the choices.
- Thinking that adding a constant to every value changes the SD, or that multiplying by -2 gives an SD of -2 times the old one. Why it tempts: The mean changes with both operations, so the same is assumed for the spread; the SD scales by the modulus of the multiplier and ignores the shift.
- Forgetting the (m_i - m)² terms when pooling the variances of two groups. Why it tempts: A weighted average of the two variances looks like the natural combined spread, and it is correct only when the group means are equal.
Time per question and fast method
Target: about 100 seconds per question.
Work with totals (sum, sum of squares) rather than deviations. Use an assumed mean and step deviation for grouped tables. For transformed data write the effect of y = a x + b directly: mean a m + b, variance a² v. For a median or mode question write the class, then l, f, cf, h in one line before substituting.
Recall quiz
Pick an option for each question. Your answers stay on this page and are not sent anywhere.
Score: 0 of 6 correct. Pick an option for each question.
Practice
- The marks of 10 students are grouped as follows: 2 students scored 10, 3 scored 20, 4 scored 30 and 1 scored… Verified (T1)
- Eight observations have mean 10 and variance 4. A ninth observation equal to 16 is added. The variance of the… Verified (T1)
- The marks of 40 students are grouped in classes of width 10: 0 to 10 has 4 students, 10 to 20 has 8, 20 to 30… Verified (T1)
- Ten observations x are transformed by y = 3x - 2. The transformed observations have mean 13 and variance 72.… Verified (T1)
- Group A has 10 observations with mean 10 and variance 9. Group B has 30 observations with mean 14 and… Verified (T3)
Previous-year questions
Previous-year questions are not imported yet. Previous-year questions are not imported yet; see the official papers on the NTA site. Official NTA papers page.
Going deeper and revision
Measures of central tendency: definitions and edge cases
Mean uses every value and is pulled by extreme values; the median ignores the size of the extremes and is the preferred centre for skewed data; the mode may not exist or may not be unique (bimodal data). For a symmetric unimodal distribution mean, median and mode coincide. For moderately skewed data the empirical relation mode = 3 median - 2 mean is used as an approximation. For grouped data the median and mode formulae assume that the class intervals are continuous and of equal width; with inclusive classes such as 10 to 19 and 20 to 29 convert to class boundaries 9.5 to 19.5 first.
Grouped data: class mark, cumulative frequency and the median class
Class mark = (lower limit + upper limit)/2. The cumulative frequency of a class is the number of observations up to and including its upper limit. The median class is the first class whose cumulative frequency is at least N/2, not the class with the highest frequency (that is the modal class, which may be different). In the median formula cf is the cumulative frequency of the class before the median class, a point where students mistakenly use the cumulative frequency of the median class itself. In the mode formula f0 and f2 are the frequencies of the classes just before and after the modal class.
Mean deviation, variance and standard deviation: formulae and units
Mean deviation about a = (sum of f |x - a|)/N. Variance = (sum of f x²)/N - (mean)², units are the square of the data unit; SD has the data unit. For n observations the variance is never negative, and it is zero only when all values are equal. For the first n natural numbers the mean is (n + 1)/2 and the variance is (n² - 1)/12. For an arithmetic progression with common difference d and n terms the variance is d² (n² - 1)/12, which follows from the previous result and the scaling rule. The coefficient of variation is (SD/mean) x 100 percent and compares the spread of two data sets with different units or means.
Linear change, combined sets and data corrections: what examiners test
The recurring test items are: (i) y = a x + b with given mean and variance of y, asking for a sum of squares of x or the mean of x; (ii) adding or removing one observation and finding the new mean and variance; (iii) combining two groups and the formula with the offset terms; (iv) correcting a wrong value found after computing the mean and SD; (v) finding an unknown frequency or value from a given mean or variance, which gives a quadratic; (vi) the minimum of the sum of squared deviations, attained at the mean, and of the sum of absolute deviations, attained at the median. The usual loss of marks is arithmetic in the sum of squares, so keep two columns, f x and f x², and total them twice.
Why variance of combined data needs the offset terms
When two groups are pooled, spread arises from within each group and from the difference between the group means. The sum of squares about the common mean m for group i equals n_i v_i + n_i (m_i - m)², since the cross term vanishes when deviations are taken about the group mean. Hence the combined variance is the weighted mean of the group variances plus the weighted mean of the squared offsets of the group means from m. If the two means are equal the offset terms vanish and the combined variance is simply the weighted mean of the variances. This is the limiting case to test any answer against.
Memory hook
Suppose a class teacher in a school in Pune looks at the marks of the 40 students in a unit test and wants to describe how the class did without reading out every mark.
- The teacher adds all marks and divides by 40 to get the average, the mean, which is the balance point of the marks.
- She lines the students up from the lowest to the highest mark and reads the mark of the student in the middle, which is the median, unaffected if the topmost mark is changed from 90 to 100.
- To say how spread out the marks are she checks how far each mark is from the average, squares those gaps and averages them: that is the variance, and its square root is the standard deviation.
Mean is the balance point, median is the middle student and standard deviation is the typical gap from the mean.
The mean balances the marks, the median is the middle student and the standard deviation is the usual distance from the average.
Revision card
Try to answer each without looking above.
- Write the median and the mode formula for grouped data and name l, cf, f, f0, f1, f2 and h.
- State variance in terms of the mean of squares and the mean, and say what divisor JEE Main uses.
- Give the effect of y = a x + b on the mean, the variance and the SD.
- State the combined-variance formula for two groups and the case in which the offset terms vanish.
- How do you correct the mean and the variance after one observation was recorded wrongly?
- About which value is the sum of absolute deviations least, and about which is the sum of squares least?
Sources and review
Reviewed Reviewed on 2026-11-20: every number recomputed by code and independently reviewed; textbook section references pending.
- Method: Every number in worked examples, recall checks and notes recomputed by code; claims recomputed independently by a separate verifier run using other numerical methods; textbook section references not yet attached; wording defects found were fixed afterwards and re-linted (code checks re-run) but not re-reviewed by a second run; syllabus scope compared with the NTA JEE Main 2026 syllabus text (by the verifier; see the report)
- Reviewer: independent verifier agent run (not the author)
- Date: 2026-11-20
- Defects found: 2; fixed: 2
- Report: content/reviews/learn4-prb-2026-11-20.md
Textbook references
- NCERT Mathematics, Class 11 (reprint 2026-27), Chapter 13: Statistics, 13.4.2 Mean deviation for grouped data (read 2026-11-22)
- NCERT Mathematics, Class 11 (reprint 2026-27), Chapter 13: Statistics, 13.5.1 Standard Deviation (read 2026-11-22)
- NCERT Mathematics, Class 11 (reprint 2026-27), Chapter 13: Statistics, 13.5 Variance and Standard Deviation (read 2026-11-22)
- NCERT Mathematics, Class 11 (reprint 2026-27), Chapter 13: Statistics, 13.5.4 Shortcut method to find variance and standard deviation (read 2026-11-22)
Sources used
- Syllabus: NTA JEE (Main) 2026 syllabus, Mathematics, Unit 13 (https://jeemain.nta.nic.in/document/syllabus-2026/), imported into syllabus/lines/jee-main/2026.json
- NTA JEE (Main) 2026 syllabus, Mathematics, Unit 13 (https://jeemain.nta.nic.in/document/syllabus-2026/), read for scope
- Standard results of school-level statistics and probability; no textbook section reference attached yet (to be sourced at verification). Content written fresh for this site.
How labels work: see how answers and lessons are verified. Found a mistake? Use Report an error on any question page.