Averages of grouped data

A cricket coach jots down the runs scored by each player over a season, a shopkeeper tallies how many customers bought each shoe size, a teacher records marks out of ten for a class quiz — in every case there is now a pile of numbers, and a single question follows: what does this data typically look like? Three familiar tools answer that question from three different angles. The mean balances every value against every other, the median picks out whoever sits exactly in the middle once everyone is lined up in order, and the mode simply names whichever value shows up most often. Class 8 does not hand you new formulas for these — you already used them on small lists. What changes here is scale: the data now arrives already counted and tabulated, with a frequency column doing the "repeating" for you, and you must read that table correctly before the same three familiar calculations go to work.

The core insight

Mean, median and mode summarise a data set from three different angles — total shared equally, the middle value in order, and the most frequent value — and a frequency table is just a shorthand for a longer list of repeated raw values, so the same three formulas apply once the table is expanded.

The core idea

Mean. For nn raw observations x1,x2,…,xnx_1, x_2, \ldots, x_n, the mean is the total shared out equally among all of them:

xˉ=x1+x2+⋯+xnn=sum of all observationsnumber of observations\bar{x} = \dfrac{x_1+x_2+\cdots+x_n}{n} = \dfrac{\text{sum of all observations}}{\text{number of observations}}

where: xˉ\bar{x} is the mean, xix_i are the individual observations, and nn is the total count of observations.

Median. Arrange every observation in ascending order first — the median only makes sense on a sorted list. Then:

Median={value at position n+12n odd12(value at n2+value at n2+1)n even\text{Median} = \begin{cases} \text{value at position } \dfrac{n+1}{2} & n \text{ odd} \\[4pt] \dfrac{1}{2}\left(\text{value at } \dfrac{n}{2} + \text{value at } \dfrac{n}{2}+1\right) & n \text{ even} \end{cases}

where: nn is the number of observations, and "position" counts along the sorted list starting from 1.

Mode. The mode is the observation that occurs most often — not the frequency itself, the value. A data set can have one mode, several tied modes, or none at all if every value occurs equally often. A frequency table does not change any of this. If marks of 6, 7 and 8 were recorded for 2, 3 and 4 students respectively, that table is just a compressed way of writing the list 6,6,7,7,7,8,8,8,86, 6, 7, 7, 7, 8, 8, 8, 8 — you can always expand a table back into a full list of raw values before applying the definitions above. This trick — repeat each value by its frequency — is the bridge between "a table of counts" and "the raw list" the mean, median and mode formulas were built for.

Figure
Three data pictures side by side showing mean as a balance point, median as the middle marker on a sorted number line, and mode as the tallest bar in a frequency chart
Mean, median and mode — three views of the same data mean — balance point total shared equally median — middle of sorted list middle value list sorted ascending first mode — tallest bar highest frequency
Derivation AMean of data recorded in a frequency table

Goal: show that when raw values are already grouped into a frequency table, the mean of the whole data set equals the sum of each value multiplied by its own frequency, divided by the total number of observations — so you never have to physically rewrite the table as one long list to find the mean.

Step 1. Suppose kk distinct values x1,x2,…,xkx_1, x_2, \ldots, x_k occur in the data, with frequencies f1,f2,…,fkf_1, f_2, \ldots, f_k respectively — meaning x1x_1 occurs f1f_1 times, x2x_2 occurs f2f_2 times, and so on. Expanded into a raw list, this data set has x1x_1 written out f1f_1 times, x2x_2 written out f2f_2 times, and so on, giving a total of n=f1+f2+⋯+fkn = f_1+f_2+\cdots+f_k observations in all.

Step 2. By the raw-data mean formula, xˉ\bar{x} is the sum of every entry in that expanded list divided by nn. The entries equal to x1x_1 contribute x1+x1+⋯+x1⏟f1 times=f1x1\underbrace{x_1+x_1+\cdots+x_1}_{f_1 \text{ times}} = f_1x_1 to the total sum, since adding the same number to itself f1f_1 times is exactly f1f_1 multiplied by that number. The same reasoning applies to every other value.

Step 3. Adding up the contribution from each distinct value gives the total sum of all observations as f1x1+f2x2+⋯+fkxkf_1x_1+f_2x_2+\cdots+f_kx_k, which is written compactly as ∑fixi\sum f_ix_i. Dividing this total by the total count n=∑fin=\sum f_i gives the mean:

xˉ=∑fixi∑fi\bar{x} = \dfrac{\sum f_i x_i}{\sum f_i}

Step 4. This is exactly the calculation used earlier: multiply each value by its own frequency, add these products, and divide by the total frequency. It is not a new formula — it is the raw-data mean formula in disguise, with the repeated-addition step replaced by a single multiplication for speed.

Worked example

The marks (out of 10) obtained by 12 students in a class quiz were recorded in a frequency table.

Marks678910
Number of students23421

Find the mean, median and mode of the marks.

Step 1. Check the total count against the given number of students: 2+3+4+2+1=122+3+4+2+1 = 12, matching the 12 students stated in the problem.

Step 2. Find the mean using xˉ=∑fixi∑fi\bar{x}=\dfrac{\sum f_ix_i}{\sum f_i}. The sum of value times frequency is (6×2)+(7×3)+(8×4)+(9×2)+(10×1)=12+21+32+18+10=93(6\times2)+(7\times3)+(8\times4)+(9\times2)+(10\times1) = 12+21+32+18+10 = 93. Dividing by the total frequency, xˉ=9312=7.75\bar{x} = \dfrac{93}{12} = 7.75 marks.

Step 3. Find the median. Expand the table into the sorted list of all 12 marks: 6,6,7,7,7,8,8,8,8,9,9,106,6,7,7,7,8,8,8,8,9,9,10. Since n=12n=12 is even, the median is the mean of the values at positions 122=6\frac{12}{2}=6 and 122+1=7\frac{12}{2}+1=7. Both the 6th and 7th values are 88, so the median is 8+82=8\dfrac{8+8}{2} = 8 marks.

Step 4. Find the mode. The mark 88 has the highest frequency in the table, 44 students, so the mode is 88 marks.

Final answer: mean =7.75=7.75, median =8=8, mode =8=8 (marks out of 10). The mean sits slightly below the median and mode because the two lowest-scoring students (marks of 6) pull the balance point down, even though most students actually scored 8 or higher.

Competitive stretch

A football team's goals in its last 11 matches were 2,0,3,1,2,4,1,2,0,3,22,0,3,1,2,4,1,2,0,3,2 — this time as a raw list, not a table. Building a quick frequency count first — 00 twice, 11 twice, 22 four times, 33 twice, 44 once — turns it into exactly the same kind of problem: mean =2011≈1.82=\dfrac{20}{11}\approx1.82, median (sorted list 0,0,1,1,2,2,2,2,3,3,40,0,1,1,2,2,2,2,3,3,4, position 11+12=6\frac{11+1}{2}=6) =2=2, mode =2=2. Recognising a raw list as an implicit frequency table, and vice versa, is the skill exam-setters test most often.

Common traps

  • Reading off the median before sorting. The median is a position in the ordered list; picking the middle entry of an unsorted table or list gives a wrong answer.
  • Forgetting to average the two middle values when nn is even — many students stop at the value nearest the middle instead of averaging positions n2\frac{n}{2} and n2+1\frac{n}{2}+1.
  • Confusing the mode with its frequency — the mode is 88 marks, not "44 students"; the frequency only tells you how often, not which value.
  • Reaching for class-interval grouped-mean methods (direct, short-cut or step-deviation, using class mid-points) — that machinery is for Class 10 grouped data with intervals like 0-100\text{-}10, 10-2010\text{-}20; at this level every frequency table lists actual individual values, so expand-and-apply is always enough.

Ask the tutor: if your mean, median or mode for a frequency table doesn't match what's expected, ask the AI tutor to walk through the expand-and-count steps with your exact numbers.

Why doesn't the mean equal the median or mode most of the time?

Because the three averages measure different things. The mean accounts for every value's exact size, so a few unusually high or low values can pull it up or down. The median only cares about position in the sorted order, ignoring how far the extreme values are from the middle. The mode ignores size entirely and just counts repeats. They coincide only for very symmetric, evenly spread data — in most real data sets they differ, and that difference is itself informative.

Do I always have to expand a frequency table into a long list before calculating anything?

No — expanding it out is a useful way to understand what the table means, but for the mean you can go straight to xˉ=∑fixi∑fi\bar{x}=\dfrac{\sum f_ix_i}{\sum f_i} without writing out every repeated value. For the median, you do need the sorted order, but you can count through the frequencies cumulatively (2, then 2+3=5, then 5+4=9, and so on) to find which value sits at the required position, instead of writing out all 12 or however many entries.

Is this the same as the grouped-mean formula I've heard about with class intervals?

No, and mixing the two up is a common Class 8 mistake. Here, every row of the frequency table is still an individual value (like 6 marks, 7 marks) with a count attached. Grouped data with class intervals (like 0-10, 10-20) and their mid-points, along with direct, short-cut or step-deviation methods for estimating the mean, is a Class 10 topic and is not tested here — at this level, expanding or applying ∑fixi/∑fi\sum f_ix_i / \sum f_i on individual values is always sufficient.

Before this

Where this leads