Measures of Dispersion for Grouped Data · Form 5
Measures of Dispersion for Grouped Data: Key Terms
The key Measures of Dispersion for Grouped Data terms you need for SPM Mathematics, defined plainly in English, Malay and Chinese.
- Ogive A cumulative frequency curve, used to read off the median and quartiles of grouped data.
- Class midpoint The middle value of a class interval, used to stand in for all values in that interval.
- Cumulative frequency A running total of frequencies, used to plot an ogive.
- Quartile A value dividing ordered data into quarters; the first, second (median) and third quartiles.
- Interpolation Reading a value between plotted points, for example finding the median from an ogive.
- Class interval A range of values into which grouped data is sorted.
- Modal class The class interval with the highest frequency in grouped data.
- Histogram A bar chart for grouped data where bar area represents frequency.
- Class midpoint The value halfway across a class interval, used to estimate the mean of grouped data.
Term pairs students confuse
- Class midpoint vs class boundary, the midpoint (e.g. 165 for 160–170) is the single value used in the mean and standard deviation; the boundaries (160 and 170) are the edges used for histogram bars and for plotting the ogive.
- Frequency vs cumulative frequency, the frequency is how many fall in one class; the cumulative frequency is the running total up to the upper boundary of that class, and it is what the ogive plots.
- Variance vs standard deviation, the variance is the mean squared deviation and is in squared units; the standard deviation is its square root, back in the data's own units, so it is the one you interpret.
- Modal class vs median class, the modal class has the highest frequency; the median class contains the n/2 th value. They can be two different classes in the same table.
- Histogram vs ogive, a histogram shows frequency as bars drawn against class boundaries; the ogive is a smooth cumulative-frequency curve used to read the median and quartiles.
- Range vs interquartile range, the range is the highest boundary minus the lowest; the interquartile range is Q3 − Q1, the spread of the middle half, which ignores the extreme classes.
How these terms show up in real SPM questions
SPM wording is a code worth decoding before you calculate. 'Estimate the mean' signals grouped data, so you must use midpoints, not a raw list.
'Using the ogive' or 'from your graph' means read the answer off the curve, you must draw it, and a working line is still expected. 'Modal class' asks for an interval such as 170–180, never a single number.
'The interquartile range' expects Q3 − Q1, read at the 3n/4 th and n/4 th positions on the cumulative axis. Instructions like 'correct to 2 decimal places' fix how you round the final standard deviation, obey them, and only at the end.
Reading these phrases first tells you which tool the marker wants.
Source:DSKP KSSM Mathematics Form 4 and 5 (Versi English)
Frequently asked questions
What is the difference between the 'modal class' and 'the mode'?
In grouped data you cannot pin down a single mode, so SPM asks for the modal class, the interval with the highest frequency, for example 170–180. A single-number mode only exists for ungrouped data.
You can estimate a mode inside the modal class from a histogram, but the term 'modal class' means the interval itself.
When a question says 'estimate the mean', why estimate rather than an exact value?
Because grouping has already discarded the individual readings. You replace each class with its midpoint and assume every member sits exactly there, so the mean you get is a good estimate rather than the exact value of the original data.
That is why SPM deliberately uses the word 'estimate' and accepts the midpoint result.
Is the 'median class' the same as the class that contains the mean?
Not necessarily. The median class is the interval holding the n/2 th value, found from the cumulative frequency.
The mean is a calculated average that may land in a different class, especially when the data is skewed. Locate each one separately and never assume the two coincide.