Skip to main content
NSWMaths Standard 2Syllabus dot point

How do you organise raw data into a frequency table, group it into class intervals with class centres, and build a cumulative frequency column?

Organise, interpret and display data into appropriate tabular and graphical representations including frequency distribution tables, both ungrouped and grouped using class intervals and class centres, and cumulative frequency

A focused answer to the HSC Maths Standard 2 dot point on frequency tables. Tallying raw data into a frequency table, grouping data into class intervals, finding the class centre, and building the cumulative frequency column, with worked Australian examples and the totals checked so the cumulative frequency ends at the sample size.

Generated by Claude Opus 4.814 min answer

Reviewed by: AI editorial process; not yet individually human-reviewed

Have a quick question? Jump to the Q&A page

What this dot point is asking

NESA wants you to take a list of raw data and organise it into a frequency table. For a small spread of values you tally each value and record how often it occurs. For data that is spread over a wide range you first group it into class intervals and work with the class centre (the middle value of each interval) as the representative figure for the class. On top of either kind of table you add a cumulative frequency column, a running total that answers "how many values are this size or smaller". None of the arithmetic is hard. The marks are won and lost on three habits: tallying without missing or double-counting a value, choosing sensible class intervals, and getting the cumulative column to end exactly at the sample size nn. Get those right and every table in the Data Analysis module becomes routine, because the histogram, the polygon and the ogive that follow are all just pictures of these columns.

The answer

A frequency table has one row per value (or per class) and a frequency column that counts how often each appears. The first habit is reliable tallying: go through the raw data once, put one stroke in the right row for each value, group the strokes in fives (four strokes with a fifth struck through them, a "gate"), then count the strokes to get each frequency. Always finish by adding the frequency column and checking it equals the number of data values nn, because a total that is off by one is the signature of a missed or double-counted value.

Building a frequency table with tallies (ungrouped data)

When the data takes only a few distinct values - shoe sizes, goals per game, number of pets - you do not need to group anything. List each distinct value in its own row from smallest to largest, tally the raw data into those rows, and write the frequency beside each. The tally column is not decoration: tallying in one careful pass is what stops you losing or repeating a value, and the gate-of-five grouping makes the strokes quick to count.

For example, suppose 2020 players report shoe sizes and the tallies come out as below. The frequency is just the number of strokes, and the column total confirms all 2020 players are accounted for.

Shoe size Tally Frequency
77 \cancel{||||} 55
88   \cancel{||||}\;||| 88
99 \cancel{||||} 55
1010 || 22

The frequencies add to 5+8+5+2=205 + 8 + 5 + 2 = 20, which matches the number of players, so nothing was missed.

Grouping raw data into class intervals

When the data is spread over a wide range - test marks out of 100100, masses in grams, commute times - a row for every distinct value would be unreadable. Instead you group the data into class intervals (also called classes), each covering an equal-width band of values, and count how many data values fall in each band.

Two choices set up a good grouped table:

  • Number of classes. Aim for roughly 55 to 1010 classes. Too few hides the shape; too many leaves the table almost as long as the raw list.
  • Class width. Pick a "nice" equal width (often 55, 1010 or 2020) so every value lands in exactly one class. Equal widths matter: they keep the class centres evenly spaced and make the later histogram fair.

Set the classes so they do not overlap and leave no gaps, for example 1-201\text{-}20, 21-4021\text{-}40, 41-6041\text{-}60 and so on. Then tally each raw value into the one class that contains it.

The class centre

Once data is grouped you no longer know each exact value, only the class it sits in. The class centre is the value used to stand in for everything in that class: it is the midpoint of the interval, found by averaging the lower and upper limit.

So the interval 41-6041\text{-}60 has class centre 41+602=50.5\dfrac{41 + 60}{2} = 50.5, and the interval 0-90\text{-}9 has class centre 0+92=4.5\dfrac{0 + 9}{2} = 4.5. Because the classes have equal width, the class centres are evenly spaced by exactly the class width, which is a fast way to check them. The diagram below shows what the class centre means on a number line.

The class centre is the midpoint of a class intervalA number line from forty to seventy. The class interval from forty-one to sixty is shaded as a band. Its lower limit forty-one and upper limit sixty are marked, and its class centre, fifty point five, sits exactly halfway between them, found by averaging the two limits.Class centre = midpoint of the interval41lower limit60upper limitclass centre 50.5(41 + 60) ÷ 2 = 50.5

The cumulative frequency column

The cumulative frequency of a row is the total of that row's frequency and all the frequencies above it - a running total down the table. It answers "how many data values are at most this size". You build it by carrying a running sum: start with the first frequency, then add each next frequency to the total so far.

The single most important check in this whole topic: the last cumulative frequency must equal nn, the number of data values, because by the bottom row you have counted everything. If it does not, a frequency is wrong or has been added twice.

Cumulative frequency is the bridge to the rest of the module. Reading "how many scored 6060 or less" off the cumulative column is exactly what the cumulative frequency graph (the ogive) does with a curve, and it is how the median and quartiles get located later.

Putting it together: raw to grouped to cumulative

The whole method is one table grown in three steps. Take the raw data, tally it into class intervals to get the frequency column, then run a total down the side to get the cumulative column, adding a class-centre column so the table is ready for the mean and the graphs that follow. The worked set below does exactly this, and every frequency, class centre and cumulative total has been checked so the cumulative column ends at nn.

How exam questions ask about frequency tables

The wording varies, but each version points straight at one of the columns:

  • "Construct / complete a frequency table" with a tally column means tally the raw data into rows and count, then check the total equals nn.
  • "Group the data into the class intervals ..." or "using a class width of ..." means build a grouped table: tally each value into its class and (usually) add a class-centre column.
  • "Find the class centre of ..." is just the average of the two class limits.
  • "State the modal class" asks for the class with the highest frequency (the grouped version of the mode).
  • "Complete the cumulative frequency column" means run a total down the frequency column; expect a check that it ends at nn.
  • "How many scored at most / less than / no more than ..." is a cumulative-frequency read. "At most" and "no more than" include the boundary value; "less than" stops below it - read the wording carefully.
  • "Find the value of the missing frequency" uses the total: subtract the known frequencies from nn.
  • "Estimate the mean" from a grouped table means use the class centres in xˉfxf\bar{x} \approx \dfrac{\sum f x}{\sum f}; the word "estimate" is the clue that grouping has hidden the exact values.

Exam-style practice questions

Practice questions written in the style of NESA exam questions on this dot point, with worked answer explainers. The year tag is the paper they imitate, not the source.

2021 HSC-style4 marksThe number of children per household in a survey of 2525 households gave the frequency table below, where one frequency is missing. | Children | Frequency | | --- | --- | | 00 | 33 | | 11 | xx | | 22 | 88 | | 33 | 55 | | 44 | 22 | (a) Find the value of xx. (b) Construct a cumulative frequency column. (c) Find the number of households with fewer than 33 children.
Show worked answer →

Part (a): the frequencies sum to the 2525 households, so x=25(3+8+5+2)=2518=7x = 25 - (3 + 8 + 5 + 2) = 25 - 18 = 7. One mark for x=7x = 7.

Part (b): cumulative frequencies are the running totals 3,10,18,23,253, 10, 18, 23, 25. One mark, and markers check the final cumulative frequency equals n=25n = 25.

Part (c): "fewer than 33 children" means 00, 11 or 22, the cumulative frequency at the 22 row, which is 3+7+8=183 + 7 + 8 = 18 households. One mark for the answer and one for correctly reading the cumulative column (not including the 33-children row).

Markers award the value of xx, a correct cumulative column ending at 2525, and the correct cumulative read for "fewer than 33".

2022 HSC-style5 marksThe examination marks of 8080 students are grouped into the class intervals 1-201\text{-}20, 21-4021\text{-}40, 41-6041\text{-}60, 61-8061\text{-}80, 81-10081\text{-}100, with frequencies 4,10,28,264, 10, 28, 26 and 1212. (a) State the class centre of the 41-6041\text{-}60 interval. (b) State the modal class. (c) Construct a cumulative frequency column. (d) A student passes by scoring more than 6060. Find how many students passed.
Show worked answer →

Part (a): the class centre is 41+602=50.5\frac{41 + 60}{2} = 50.5. One mark.

Part (b): the modal class is 41-6041\text{-}60 because it has the highest frequency, 2828. One mark.

Part (c): the cumulative frequencies are 4,14,42,68,804, 14, 42, 68, 80, ending at the total 8080. One mark, with the final value checked against n=80n = 80.

Part (d): "more than 6060" means the 61-8061\text{-}80 and 81-10081\text{-}100 classes, so 26+12=3826 + 12 = 38 students passed. Equivalently, the cumulative frequency up to 41-6041\text{-}60 is 4242, and 8042=3880 - 42 = 38. One mark for the method, one for 3838.

Markers reward the class centre, the modal class identified by frequency, a cumulative column ending at 8080, and the correct "more than 6060" count.

2023 HSC-style3 marksA grouped frequency table records the masses of 4040 parcels in classes of width 55 kg starting at 0-40\text{-}4. The frequencies of the first four classes are 6,11,146, 11, 14 and 99. (a) State the upper class limit and the class centre of the second class (5-95\text{-}9). (b) Show that these four classes account for all 4040 parcels and write the cumulative frequency of the fourth class.
Show worked answer →

Part (a): the second class is 5-95\text{-}9, so its upper class limit is 99 and its class centre is 5+92=7\frac{5 + 9}{2} = 7. One mark for both.

Part (b): the four frequencies sum to 6+11+14+9=406 + 11 + 14 + 9 = 40, the total number of parcels, so the four classes account for every parcel. The cumulative frequency of the fourth class is this running total, 4040. One mark for the sum equalling 4040 and one for stating the cumulative frequency is 4040 (the total).

Markers reward the correct upper limit and class centre, the demonstration that the frequencies total 4040, and the cumulative frequency stated as 4040.

Practice questions

Original practice questions graded from foundation to exam level, each with a full worked solution. Try them before revealing the solution.

foundation2 marksA frequency table records the number of pets owned by each of 3030 students. The frequencies for 0,1,2,30, 1, 2, 3 and 44 pets are 6,11,8,46, 11, 8, 4 and 11. (a) Confirm the frequencies add to the number of students. (b) State how many students own at least 22 pets.
Show worked solution →

Part (a) - add the frequency column. The frequency column should total the number of students surveyed:

6+11+8+4+1=306 + 11 + 8 + 4 + 1 = 30

This matches the 3030 students, so the table is complete (every student is counted exactly once).

Part (b) - "at least 22" means 22 or more. Add the frequencies for 22, 33 and 44 pets:

8+4+1=138 + 4 + 1 = 13

so 1313 students own at least 22 pets. (Sanity check: 1313 is less than the total 3030, as it must be, and the 1717 students with 00 or 11 pet make up the rest, since 6+11=176 + 11 = 17 and 17+13=3017 + 13 = 30.)

foundation3 marksTwenty netball players are asked their shoe size. The results are 8,7,9,8,10,7,8,9,8,7,9,8,10,8,7,9,8,8,7,98, 7, 9, 8, 10, 7, 8, 9, 8, 7, 9, 8, 10, 8, 7, 9, 8, 8, 7, 9. Construct a frequency table with a tally column for the shoe sizes.
Show worked solution →

Set up the rows. List each distinct shoe size from smallest to largest in its own row: 7,8,9,107, 8, 9, 10.

Tally each value, then count. Work through the list once, adding one stroke per value (grouping in fives with a gate stroke), then write the frequency as the number of strokes:

Shoe size Tally Frequency
77 \cancel{||||} 55
88   \cancel{||||}\;||| 88
99 \cancel{||||} 55
1010 || 22

Check the total. The frequency column must add to the 2020 players:

5+8+5+2=205 + 8 + 5 + 2 = 20

which matches, so no value has been missed or double-counted.

foundation2 marksA grouped frequency table uses the class intervals 1-101\text{-}10, 11-2011\text{-}20, 21-3021\text{-}30 and 31-4031\text{-}40. (a) Find the class centre of each interval. (b) State which interval the value 2424 falls into.
Show worked solution →

Part (a) - the class centre is the average of the two endpoints. Add the lower and upper limit of each interval and halve:

1+102=5.5,11+202=15.5,21+302=25.5,31+402=35.5\frac{1 + 10}{2} = 5.5, \quad \frac{11 + 20}{2} = 15.5, \quad \frac{21 + 30}{2} = 25.5, \quad \frac{31 + 40}{2} = 35.5

so the class centres are 5.5,15.5,25.55.5, 15.5, 25.5 and 35.535.5.

Part (b) - locate the value. Since 21243021 \le 24 \le 30, the value 2424 falls in the interval 21-3021\text{-}30. (Note the class centres are evenly spaced 1010 apart, which is the class width - a quick check that they are right.)

core4 marksThe heights of 2525 tomato seedlings, in centimetres, are recorded: 12,7,23,18,5,29,14,21,9,33,16,25,11,38,20,8,27,15,31,22,13,19,24,10,1712, 7, 23, 18, 5, 29, 14, 21, 9, 33, 16, 25, 11, 38, 20, 8, 27, 15, 31, 22, 13, 19, 24, 10, 17. Group the data into the class intervals 0-90\text{-}9, 10-1910\text{-}19, 20-2920\text{-}29 and 30-3930\text{-}39, and build a grouped frequency table with a class-centre column and a frequency column.
Show worked solution →

Decide the rows from the given intervals. The four classes are 0-90\text{-}9, 10-1910\text{-}19, 20-2920\text{-}29 and 30-3930\text{-}39, each 1010 wide.

Tally each value into its class, then count the frequency. Going through the list once and dropping each height into the interval that contains it gives the frequencies below. The class centre of each interval is the average of its endpoints, for example 0+92=4.5\dfrac{0 + 9}{2} = 4.5.

Height (cm) Class centre Frequency
0-90\text{-}9 4.54.5 44
10-1910\text{-}19 14.514.5 1010
20-2920\text{-}29 24.524.5 88
30-3930\text{-}39 34.534.5 33

Check the total. The frequencies must add to the 2525 seedlings:

4+10+8+3=254 + 10 + 8 + 3 = 25

which matches, so every seedling has been placed in exactly one class. (The class centres 4.5,14.5,24.5,34.54.5, 14.5, 24.5, 34.5 rise by 1010, the class width, confirming they are evenly spaced.)

core4 marksThe number of goals scored by a hockey team in 3030 matches gave the frequencies below. Add a cumulative frequency column, and use it to find how many matches the team scored at most 22 goals in. | Goals | Frequency | | --- | --- | | 00 | 22 | | 11 | 55 | | 22 | 99 | | 33 | 88 | | 44 | 44 | | 55 | 22 |
Show worked solution →

Build the cumulative frequency by running totals. Each cumulative frequency is the sum of all frequencies up to and including that row, so add each new frequency to the running total:

Goals Frequency Cumulative frequency
00 22 22
11 55 2+5=72 + 5 = 7
22 99 7+9=167 + 9 = 16
33 88 16+8=2416 + 8 = 24
44 44 24+4=2824 + 4 = 28
55 22 28+2=3028 + 2 = 30

Check the final cumulative total. The last cumulative frequency must equal the number of matches, n=30n = 30, and it does (28+2=3028 + 2 = 30), so the column is correct.

Read off "at most 22 goals". "At most 22" means 00, 11 or 22 goals, which is exactly the cumulative frequency in the 22 row:

2+5+9=162 + 5 + 9 = 16

so the team scored at most 22 goals in 1616 matches.

exam6 marksThe daily commuting times, in minutes, of 5050 workers are recorded: 5,12,23,34,8,15,27,41,19,22,7,18,29,33,11,25,38,44,16,21,3,14,26,31,9,20,37,42,17,24,6,13,28,35,10,23,39,45,18,20,4,16,24,32,12,21,30,43,15,225, 12, 23, 34, 8, 15, 27, 41, 19, 22, 7, 18, 29, 33, 11, 25, 38, 44, 16, 21, 3, 14, 26, 31, 9, 20, 37, 42, 17, 24, 6, 13, 28, 35, 10, 23, 39, 45, 18, 20, 4, 16, 24, 32, 12, 21, 30, 43, 15, 22. (a) Group the data into class intervals 0-90\text{-}9, 10-1910\text{-}19, 20-2920\text{-}29, 30-3930\text{-}39, 40-4940\text{-}49 and build a grouped frequency table with class-centre, frequency and cumulative frequency columns. (b) State the modal class. (c) Find how many workers commute less than 3030 minutes. (d) Estimate the mean commuting time using the class centres.
Show worked solution →

Part (a) - tally into classes, then add the class-centre and cumulative columns. Each value is placed in the one class that contains it; the class centre is the average of the endpoints (for example 0+92=4.5\dfrac{0 + 9}{2} = 4.5); and each cumulative frequency is the running total of the frequencies.

Time (min) Class centre xx Frequency ff Cumulative frequency
0-90\text{-}9 4.54.5 77 77
10-1910\text{-}19 14.514.5 1414 2121
20-2920\text{-}29 24.524.5 1515 3636
30-3930\text{-}39 34.534.5 99 4545
40-4940\text{-}49 44.544.5 55 5050

The frequencies add to 7+14+15+9+5=507 + 14 + 15 + 9 + 5 = 50, and the final cumulative frequency is 50=n50 = n, so the table is consistent.

Part (b) - modal class. The highest frequency is 1515, so the modal class is 20-2920\text{-}29 minutes.

Part (c) - "less than 3030 minutes". Times under 3030 fall in the classes 0-90\text{-}9, 10-1910\text{-}19 and 20-2920\text{-}29, which is the cumulative frequency at the 20-2920\text{-}29 row:

7+14+15=367 + 14 + 15 = 36

so 3636 workers commute less than 3030 minutes.

Part (d) - estimate the mean from class centres. With grouped data the mean is estimated by treating every value in a class as its class centre, so multiply each class centre by its frequency, add, and divide by nn:

fx=4.5(7)+14.5(14)+24.5(15)+34.5(9)+44.5(5)\sum f x = 4.5(7) + 14.5(14) + 24.5(15) + 34.5(9) + 44.5(5)

fx=31.5+203+367.5+310.5+222.5=1135\sum f x = 31.5 + 203 + 367.5 + 310.5 + 222.5 = 1135

xˉfxf=113550=22.7 minutes\bar{x} \approx \frac{\sum f x}{\sum f} = \frac{1135}{50} = 22.7 \text{ minutes}

so the estimated mean commuting time is about 22.722.7 minutes. (It is an estimate, not the exact mean, because the class centres stand in for the real values once the data is grouped.)

exam5 marksThe masses, in grams, of 6060 apples are grouped into the class intervals 100-109100\text{-}109, 110-119110\text{-}119, 120-129120\text{-}129, 130-139130\text{-}139 and 140-149140\text{-}149, giving frequencies 8,15,18,118, 15, 18, 11 and 88. (a) Find the class centre of each interval. (b) Build a cumulative frequency column. (c) An apple is graded 'large' if its mass is 130130 g or more. Estimate how many of the 6060 apples are large, and express this as a percentage of the sample.
Show worked solution →

Part (a) - class centres. The class centre is the average of the lower and upper limit of each interval:

100+1092=104.5,110+1192=114.5,120+1292=124.5,\frac{100 + 109}{2} = 104.5, \quad \frac{110 + 119}{2} = 114.5, \quad \frac{120 + 129}{2} = 124.5,

130+1392=134.5,140+1492=144.5\frac{130 + 139}{2} = 134.5, \quad \frac{140 + 149}{2} = 144.5

so the class centres are 104.5,114.5,124.5,134.5104.5, 114.5, 124.5, 134.5 and 144.5144.5 grams.

Part (b) - cumulative frequency column. Add the frequencies as a running total:

Mass (g) Class centre Frequency Cumulative frequency
100-109100\text{-}109 104.5104.5 88 88
110-119110\text{-}119 114.5114.5 1515 2323
120-129120\text{-}129 124.5124.5 1818 4141
130-139130\text{-}139 134.5134.5 1111 5252
140-149140\text{-}149 144.5144.5 88 6060

The final cumulative frequency is 60=n60 = n, confirming the column.

Part (c) - "large" apples are 130130 g or more. These are the two top classes, 130-139130\text{-}139 and 140-149140\text{-}149:

11+8=19 apples11 + 8 = 19 \text{ apples}

As a percentage of the 6060 apples:

1960×10031.7%\frac{19}{60} \times 100 \approx 31.7\%

so about 1919 apples (roughly 31.7%31.7\%) are large. (Cross-check with the cumulative column: 4141 apples are under 130130 g, and 6041=1960 - 41 = 19 are 130130 g or more, which agrees.)

Related dot points