Descriptive Statistics AA SL
Descriptive statistics summarise a data set with a few numbers and a few diagrams: where the data is centred, how spread out it is, and what its shape looks like. This page shows how to find and choose between the mean, median and mode, how to measure spread with the range, interquartile range and standard deviation, and how to read histograms, cumulative frequency graphs and box plots, with worked examples and exam-style questions.
Formulas at a glance
The mean of a frequency table and the interquartile range are in the formula booklet. The outlier boundaries, the effect of a constant change on the data and the midpoint rule for grouped data are not, so learn them. Standard deviation and variance come from the calculator.
Spread of the middle 50%
In the bookletIdentifying potential outliers
Outlier boundary: ,
Not in the bookletMean of grouped/frequency data
In the bookletWhich measure or display to use
Choose the measure from the shape of the data and what the question needs.
| The question asks for | Use | Notes |
|---|---|---|
| A typical value for data with no outliers | The mean | Enter the data as a list on the calculator. Pair it with the standard deviation |
| A typical value for skewed data or data with outliers | The median | It depends on position only, so extreme values do not move it. Pair it with the interquartile range |
| The mean from a frequency table, or an estimate of the mean from grouped data | , using the midpoint of each class for grouped data | Grouped data gives an estimate only. The modal class is the class with the highest frequency |
| The spread of the middle half of the data | The interquartile range | Not affected by outliers, unlike the range |
| The spread of all the data | The standard deviation , from the calculator | Use on the calculator screen. It is never negative |
| The variance | The square of the standard deviation | If the variance is |
| The effect of adding to every value or multiplying every value by | Adding changes the mean and median only. Multiplying by changes the mean, median and spread | Standard deviation and interquartile range are multiplied by ; adding a constant leaves them unchanged |
| Whether a value is an outlier | Below or above | Shown as a cross on a box plot, outside the whiskers |
| The median or quartiles from a cumulative frequency graph | Read across at , or on the vertical axis, then down to the horizontal axis | Plot each cumulative frequency at the upper boundary of its class |
| A comparison of two data sets | Compare a measure of centre and a measure of spread, in context | For example, the median and the interquartile range read from two box plots |
Section 1 of 3 SL4.3
Measures of central tendency
The mean, median and mode are three measures of the centre of a data set: the mean is the total divided by the number of values, the median is the middle value of the ordered data, and the mode is the most frequent value. Use the median when extreme values would distort the mean. For a frequency table, weight each value by its frequency.
The key methods
Median (ordered data)
Middle value if is odd; mean of the two middle values if is even.
For grouped data, the median is estimated from a cumulative frequency graph or by interpolation within the class that contains it.
Key terms
What is the modal class?
For grouped data, the modal class is the class interval with the highest frequency. It is used because the individual values are not known.
e.g. Classes , and with frequencies , and have modal class .
How do you estimate the mean of grouped data?
Use the midpoint of each class as its value, then calculate with these midpoints. The result is only an estimate because the true values are unknown.
e.g. Classes and with frequencies and have midpoints and , so the estimated mean is .
How do you find the median of an even number of values?
Order the data, then take the mean of the two middle values. The median is the value halfway through the ordered list.
e.g. For , the middle values are and , so the median is .
Why is the mean affected by extreme values?
The mean uses every value in its total, so one extreme value pulls it towards that value. The median depends only on position, so it is much less affected.
e.g. For the mean and median are both ; replacing with changes the mean to but leaves the median at .
Worked example, step by step
The question No calculator needed [5 marks]
The lengths of leaves are summarised in the table.
(a) Write down the modal class.
(b) Estimate the mean length of a leaf.
Worked solution: worked example
Answer: (a) The modal class is . (b) The estimated mean length is mm.
Step 1 - Find the modal class.
The highest frequency is , so the modal class is . A1 - Modal class
Why? The mode of grouped data cannot be found exactly, because the individual values are not known. The modal class is the class with the largest frequency.
Step 2 - Find the midpoint of each class.
M1 - Use the midpoints
Why? The individual lengths are not known, so each leaf in a class is treated as if its length were the midpoint of the class. This is why the answer is only an estimate.
Step 3 - Find the total of the lengths.
A1 -
Why? Each midpoint is repeated as many times as the frequency, so multiply each midpoint by its frequency before adding. This is the same weighted total you would use for a frequency table of exact values.
Step 4 - Divide by the total frequency.
M1 - Divide by A1 -
Why? The mean is the total of the lengths divided by the number of leaves. Divide by the total frequency , not by the number of classes.
Step 5 - Check.
The estimate lies between the smallest and largest midpoints, and , and is close to the middle of the table, which is where most of the leaves are.
Why? A mean must always lie within the range of the data. Dividing by instead of is a common slip, and it gives an answer that fails this check.
A1 Modal class
M1 Use the midpoints
A1
M1 Divide by
A1
Now you try: enter your answers to be marked
These are real practice questions. Type your answer in each box and press Mark it. The worked solution and mark scheme appear straight away, and your first attempt is the one that counts. Use the keyboard for powers, roots and trig, and the working box for your notes.
| Pets | 0 | 1 | 2 | 3 | 4 |
|---|---|---|---|---|---|
| Families | 8 | 10 | 7 | 4 | 2 |
Worked solution: practice question 1
| Value | 2 | 4 | 6 | 8 |
|---|---|---|---|---|
| Frequency | 3 | 5 | 2 |
The mean is 5.
Find .
Worked solution: practice question 2
Worked solution: practice question 3
Your score so far: 0 out of 11 marks
Calculator steps: One-variable statistics (mean, median, standard deviation)
Common mistakes
- Averaging the distinct values instead of weighting by frequency. For a table with values and frequencies , averaging the four values gives , but the correct mean is , because the frequencies have to multiply into the total first.
- Forgetting to reorder the data before finding the median. The median only works on ordered data - if the list is given in the order it was collected, sort it first or the "middle" value is meaningless.
- Assuming every data set has a mode. If every value occurs the same number of times there is no mode, and if two values tie for most frequent the set is bimodal - don't force a single answer where none exists.
Section 2 of 3 SL4.3
Measures of spread
Measures of spread describe how far a data set is scattered around its centre. The range uses only the extreme values, the interquartile range measures the middle half of the data, and the standard deviation uses every value. Use a measure of spread alongside the mean or median to compare two data sets.
The key measures
Effect of a linear transformation
If , the new mean is and the new standard deviation is .
Adding or subtracting a constant shifts every value equally, so it never changes the spread - only the multiplier does.
Key terms
What is variance?
Variance is the square of the standard deviation, so it is measured in squared units. A larger variance means the data is more spread out.
e.g. If the standard deviation is , the variance is .
What are quartiles?
The lower quartile , the median and the upper quartile split ordered data into four equal parts. About 25% of the data lies below and about 75% lies below .
e.g. For , the median is , and .
What does standard deviation measure?
The standard deviation measures the typical distance of the values from the mean. A smaller value means the data is more consistent, and a larger value means more variation.
e.g. The sets and both have mean , but the second has the larger standard deviation.
How does a constant change affect the spread?
Adding a constant to every value shifts the data but leaves the standard deviation and interquartile range unchanged. Multiplying every value by a positive constant multiplies both by .
e.g. If the data has standard deviation , adding to each value keeps it at , and doubling each value changes it to .
Worked example, step by step
The question Calculator needed (a list and one-variable statistics) [6 marks]
The times, in minutes, taken by people to complete a puzzle are
(a) Find the mean and the standard deviation of the times.
(b) A second puzzle takes each person twice as long as the first, plus minutes. Write down the mean and the standard deviation of the times for the second puzzle.
Worked solution: worked example
Answer: (a) Mean minutes, standard deviation minutes. (b) Mean minutes, standard deviation minutes.
Step 1 - Enter the data and read the mean.
Put the eight times in a list and run one-variable statistics. The screen shows . Check: and . M1 - Use a list and one-variable statistics, or , for the mean A1 -
Why? The calculator gives the mean directly from the list, so you do not need to add the values yourself. Adding them is worth doing once as a check that the list was entered correctly.
Step 2 - Read the standard deviation.
The screen also shows , so the standard deviation is minutes to significant figures. A1 -
Why? Use the value labelled for the standard deviation of the data you are given. Keep the unrounded value on the calculator if you use it again.
Step 3 - Decide how the change affects each measure.
Each new time is . Multiplying every value by multiplies the mean and the standard deviation by . Adding increases the mean by but leaves the standard deviation unchanged. M1 - Mean multiplied by and increased by , standard deviation multiplied by only
Why? Adding a constant slides every value along by the same amount, so the values stay the same distance apart and the spread does not change. Multiplying stretches the distances between the values, so the spread is multiplied too.
Step 4 - New mean.
A1 - New mean
Why? The mean follows the same rule as every value: double it, then add .
Step 5 - New standard deviation.
A1 - New standard deviation
Why? Only the multiplier acts on the standard deviation, and the has no effect. You can check by running one-variable statistics on the new list .
M1 Use a list and one-variable statistics, or , for the mean
A1
A1
M1 Mean multiplied by and increased by , standard deviation multiplied by only
A1 New mean
A1 New standard deviation
Now you try: enter your answers to be marked
Worked solution: practice question 1
| Goals | 0 | 1 | 2 | 3 | 4 |
|---|---|---|---|---|---|
| Freq | 6 | 14 | 11 | 6 | 3 |
Worked solution: practice question 2
Worked solution: practice question 3
Your score so far: 0 out of 12 marks
Calculator steps: One-variable statistics (mean, median, standard deviation), Box-and-whisker plot and quartiles
Common mistakes
- Reading the wrong standard deviation off the GDC. The calculator's 1-Var Stats screen shows two standard deviations. Use the one labelled for a data set you are given in full, and do not use .
- Forgetting that shifting data doesn't change its spread. Under , the constant moves every value by the same amount, so gaps between values - and hence the standard deviation - are unaffected; only rescales the spread.
- Splitting an odd-sized data set incorrectly for the quartiles. When is odd, whether the median itself is included in the lower and upper halves changes and . The calculator and the hand method can give slightly different quartiles, so follow the method the question asks for.
Section 3 of 3 SL4.2
Statistical diagrams
Histograms, cumulative frequency graphs and box and whisker diagrams are three ways of presenting a data set. A histogram shows how the data is distributed, a cumulative frequency graph is used to read the median, quartiles and percentiles, and a box plot summarises five key values and makes two data sets easy to compare.
Reading the diagrams
Box and whisker diagrams
A box plot displays the five-number summary: minimum, , median, , maximum. The box spans the IQR, and the whiskers extend to the most extreme values that aren't outliers.
Outliers are plotted as separate crosses or dots, not included in the whisker.
Cumulative frequency graph
Plots running totals against the upper class boundary; the curve rises from to .
The median is read at cumulative frequency ; at and at .
Key terms
What is cumulative frequency?
Cumulative frequency is the running total of the frequencies up to the upper boundary of each class. The final value equals the total frequency.
e.g. Frequencies give cumulative frequencies .
What is a five-number summary?
It is the minimum, lower quartile, median, upper quartile and maximum. These five values are exactly what a box and whisker diagram displays.
e.g. For , the summary is .
How do you read the interquartile range from a cumulative frequency graph?
Read the lower quartile at one quarter of the total frequency and the upper quartile at three quarters, then subtract: .
e.g. A total frequency of means reading at and on the vertical axis; readings of and give an interquartile range of .
What is an outlier on a box plot?
An outlier is a value that lies far from the rest of the data. In IB Mathematics it is a value more than above or below , and it is marked separately from the whiskers.
e.g. If and , then , so values above or below are outliers.
Worked example, step by step
The question No calculator needed [6 marks]
The times, in minutes, that students spent on an online quiz are, in order,
(a) Find the lower quartile and the upper quartile.
(b) Use the interquartile range rule to find the boundaries beyond which a value is an outlier.
(c) Hence state which times are outliers, and write down the ends of the whiskers of a box plot of the data.
Worked solution: worked example
Answer: (a) and . (b) The boundaries are and . (c) The time is the only outlier. The whiskers end at and .
Step 1 - Find the median and split the data.
There are values, so the median is the th value, . Leave the median out and split the rest into two halves of five values: M1 - Split the ordered data into two halves, leaving out the median
Why? The quartiles are the medians of the lower half and the upper half. With an odd number of values the median itself belongs to neither half, which is why it is left out here.
Step 2 - Read the quartiles.
The middle value of the lower half is the rd, so . The middle value of the upper half is , so . A1 - and
Why? Each half has values, so its middle value is the rd. A calculator can also give the quartiles, but the method used for them can differ from the method by hand, and for these data the two agree.
Step 3 - Find the interquartile range and the boundaries.
M1 - Use beyond the quartiles A1 - Boundaries and
Why? A value is an outlier if it is more than below or above . Work out both boundaries before you look at the data.
Step 4 - Decide which values are outliers.
No time is below , since the smallest is . The time is above , so is an outlier. No other time is above . R1 - Compare with the boundaries: is the only outlier
Why? Check every value against both boundaries and say what you find, including that nothing falls below the lower boundary. The conclusion is only complete when each end has been dealt with.
Step 5 - Describe the box plot.
The upper whisker ends at the largest value that is not an outlier, which is . The lower whisker ends at the smallest value, . The outlier is shown with a cross. A1 - Whiskers end at and
Why? A whisker stops at the most extreme value that is not an outlier, not at the boundary itself. Outliers are marked separately with a cross.
M1 Split the ordered data into two halves, leaving out the median
A1 and
M1 Use beyond the quartiles
A1 Boundaries and
R1 Compare with the boundaries: is the only outlier
A1 Whiskers end at and
Now you try: enter your answers to be marked
| Number of siblings | 0 | 1 | 2 | 3 | 4 |
|---|---|---|---|---|---|
| Frequency | 3 | 7 | 6 | 3 | 1 |
Worked solution: practice question 1
Worked solution: practice question 2
| Length of call (minutes) | t < 5 | t < 10 | t < 15 | t < 20 | t < 25 | t < 30 |
|---|---|---|---|---|---|---|
| Cumulative frequency | 16 | 60 | 110 | 158 | 186 | 200 |
Worked solution: practice question 3
Your score so far: 0 out of 17 marks
Calculator steps: Box-and-whisker plot and quartiles, Histogram and cumulative frequency graph
Common mistakes
- Reading the median at the wrong cumulative frequency. The median sits at on the cumulative frequency axis, not at the middle of the -axis - always work from the vertical axis first, then read across and down.
- Judging skewness from the whiskers instead of the box. Skewness is about which side of the median the box (the middle 50%) is more stretched, not which whisker looks longer - a long whisker can exist on either side regardless of the overall skew.
- Including outliers in the whisker length. A box plot's whiskers should stop at the most extreme non-outlier value; genuine outliers are marked separately, so don't stretch a whisker out to catch them.
Using your GDC
The exact key presses for the TI-84 Plus CE, Casio fx-CG50, Casio fx-CG100 and TI-Nspire CX, with what the screen shows. Pick your calculator.
One-variable statistics (mean, median, standard deviation)
Used in: Measures of central tendency, Measures of spread
How to find summary statistics on a TI-84 Plus CE
1-Var Stats (from a list) | |
Press stat and choose 1:Edit. If L1 already has numbers in it, press ▲ to put the cursor on the name L1, then press clear and enter to empty the list. |
|
Type the data into L1, one value at a time, pressing enter after each: 4, 5, 5, 6, 7, 8, 9, 10. |
L1 4 5 5 6 7 8 9 10 |
Press stat, then ▶ to move to CALC, and choose 1:1-Var Stats. With the Stat Wizards on, set Xlist to L1 (2nd 1), leave FreqList empty, move down to Calculate and press enter. If the Stat Wizards are off, 1-Var Stats is pasted on the home screen: press 2nd 1 and then enter. |
1-Var Stats Xlist:L1 FreqList: Calculate |
The results appear. Press ▼ to scroll down to minX, Q1, Med, Q3 and maxX. x̄ is the mean, σx is the standard deviation to use, and Sx is not needed. Mean and standard deviation , each to 3 significant figures. |
x̄=6.75 Σx=54 Σx²=396 Sx=2.121320 σx=1.984313 n=8 minX=4 Q1=5 Med=6.5 Q3=8.5 maxX=10 |
Data with frequencies | |
For a frequency table, type the values in L1 and the frequencies in L2. In 1-Var Stats set Xlist to L1 and FreqList to L2 (2nd 2). For the table in the box below, L1 is 1, 2, 3, 4, 5 and L2 is 3, 5, 8, 4, 2. |
1-Var Stats Xlist:L1 FreqList:L2 Calculate |
Use σx for the standard deviation and ignore Sx. The calculator's Q1 and Q3 are the medians of the lower and upper halves of the data.
The full guide: One-variable statistics (mean, median, standard deviation)
Box-and-whisker plot and quartiles
Used in: Measures of spread, Statistical diagrams
How to find quartiles and draw a box-and-whisker plot on a TI-84 Plus CE
1-Var Stats (the five-number summary) | |
Press stat and choose 1:Edit.... Type the data into L1, pressing enter after each value: 12, 6, 31, 9, 15, 5, 10, 13, 6, 3, 16, 8. If L1 already holds numbers, clear them first. |
L1 12 6 31 9 15 ... |
Press stat, then ▶ to move to CALC, and choose 1:1-Var Stats. With the Stat Wizards on, set Xlist to L1 by pressing 2nd 1 (L1), leave FreqList empty, then choose Calculate and press enter. If the Stat Wizards are off, type |
1-Var Stats Xlist:L1 FreqList: Calculate |
Press ▼ to scroll past the mean and the standard deviations until you reach minX, Q1, Med, Q3 and maxX. Minimum , , median , , maximum . So . |
n=12 minX=3 Q1=6 Med=9.5 Q3=14 maxX=31 |
Box plot | |
Press 2nd Y= (STAT PLOT) and choose 1:Plot1. Set Plot1 to On, choose the modified box plot icon for Type (the fourth of the six; it plots outliers as separate points), set Xlist to L1 with 2nd 1, and set Freq to 1. |
Plot1 On Type: modified box plot Xlist:L1 Freq:1 |
Press zoom and choose 9:ZoomStat. The box plot appears with the window set for you. Press trace: the cursor starts at the median, ◀ moves to Q1 and the lower whisker end, and ▶ moves to Q3, the upper whisker end and the outlier. The upper whisker ends at and is plotted on its own, as an outlier. |
Med=9.5 x=16 x=31 |
If the plot does not show, turn off any equations in Y= and check that only the plot you set up is on. The fifth icon is the plain box plot, whose whiskers always reach the minimum and maximum.
The full guide: Box-and-whisker plot and quartiles
Histogram and cumulative frequency graph
Used in: Statistical diagrams
How to draw a histogram and a cumulative frequency graph on a TI-84 Plus CE
Histogram from grouped data | |
Press stat and choose 1:Edit.... In L1 type the class mid-values, pressing enter after each: 5, 15, 25, 35, 45, 55. In L2 type the frequencies: 3, 7, 12, 10, 6, 2. Clear any old numbers out of both lists first. |
L1 L2 5 3 15 7 25 12 35 10 ... |
Press 2nd Y= (STAT PLOT) and choose 1:Plot1. Set Plot1 to On, choose the histogram icon for Type (the third of the six), set Xlist to L1 with 2nd 1, and set Freq to L2 with 2nd 2. |
Plot1 On Type: histogram Xlist:L1 Freq:L2 |
Press window and set Xmin to 0, Xmax to 60 and Xscl to the class width, 10. Set Ymin to 0 and Ymax to a little more than the greatest frequency, such as 14. Press graph. |
Xmin=0 Xmax=60 Xscl=10 Ymin=0 Ymax=14 |
Press trace and use ◀ and ▶ to move from bar to bar. The screen shows the start and end of each class and its frequency. The tallest bar is the class , with frequency . |
min=20 max=30 n=12 |
Cumulative frequency graph | |
Press stat and choose 1:Edit.... In L3 type the upper class boundaries: 10, 20, 30, 40, 50, 60. In L4 type the cumulative frequencies, the running totals of the frequencies: 3, 10, 22, 32, 38, 40. |
L3 L4 10 3 20 10 30 22 40 32 ... |
Press 2nd Y= (STAT PLOT), choose 1:Plot1 and set it to Off. Choose 2:Plot2, set it to On, and choose the xyLine icon for Type (the second of the six). Set Xlist to L3 with 2nd 3 and Ylist to L4 with 2nd 4. |
Plot1 Off Plot2 On Type: xyLine Xlist:L3 Ylist:L4 |
Press window and set Xmin to 0, Xmax to 60, Xscl to 10, Ymin to 0 and Ymax to a little more than the total, such as 45. Press graph. The points are joined by straight lines. Cumulative frequencies . The median is the th value, so minutes. |
xyLine (10,3) (20,10) ...(60,40) |
The graph also passes through (0, 0), the start of the first class, which the plot above leaves out: add it when you copy the graph. To avoid adding the frequencies by hand, press 2nd stat (LIST), choose OPS and then 6:cumSum(, and store cumSum(L2) in L4.
The full guide: Histogram and cumulative frequency graph
See the full list of GDC guides for every calculator task.
Syllabus content covered
| Code | Syllabus content |
|---|---|
| SL4.1 | Interpretation of outliers. An outlier is defined as a data item which is more than 1.5 × interquartile range (IQR) from the nearest quartile. Awareness that, in context, some outliers are a valid part of the data but some outlying data items may be an error. |
| SL4.2 | Presentation of data (discrete and continuous): frequency distributions (tables). Histograms. Cumulative frequency; cumulative frequency graphs; use to find median, quartiles, percentiles, range and interquartile range (IQR). Production and understanding of box and whisker diagrams. Class intervals will be given as inequalities, without gaps. Frequency histograms with equal class intervals. Use of box and whisker diagrams to compare two distributions, using symmetry, median, interquartile range or range. Outliers should be indicated with a cross. Determining whether the data may be normally distributed by consideration of the symmetry of the box and whiskers. |
| SL4.3 | Measures of central tendency (mean, median and mode). Estimation of mean from grouped data. Modal class. Measures of dispersion (interquartile range, standard deviation and variance). Effect of constant changes on the original data. Quartiles of discrete data. Calculation of mean using formula and technology. Students should use mid-interval values to estimate the mean of grouped data, for equal class intervals only. Calculation of standard deviation and variance using only technology, however hand calculations may enhance understanding. Variance is the square of the standard deviation. Examples: if three is subtracted from the data items, then the mean is decreased by three, but the standard deviation is unchanged; if all the data items are doubled, the mean is doubled and the standard deviation is also doubled. Quartiles are found using technology; the values obtained using technology and by hand may differ. |
Official IB source: Mathematics: analysis and approaches subject brief (PDF, ibo.org).
Quick questions
How do you find the mean from a frequency table?
Use the weighted mean : multiply each value by its frequency, add the results, then divide by the total frequency. On a GDC, enter values and frequencies into two lists and run 1-Variable Statistics with a frequency list.
What is the difference between the range and the interquartile range?
The range is the maximum minus the minimum, so it uses only the two most extreme values. The interquartile range, , is the spread of the middle 50% of the data, so it isn't distorted by outliers the way the range is.
How do you find the median from a cumulative frequency graph?
Find on the cumulative frequency axis, draw a horizontal line across to the curve, then a vertical line down to the horizontal axis - that value is the estimated median.
When should I use the median instead of the mean?
Use the median when the data has extreme values or is strongly skewed, because it is not pulled towards them. The mean suits data that is roughly symmetric, where every value should count.
Why is the standard deviation never negative?
It is the square root of the variance, which is an average of squared distances, so it is zero or positive. It equals only when every value is identical.
Why do the bars of a histogram touch?
Continuous data has no gaps between classes, so the bars meet. Each bar covers one class interval, and with equal class widths its height shows the frequency.
How do I find the median from a frequency table?
Find the total frequency , then use the cumulative frequencies to locate the middle position. For values with frequencies , the total is and the cumulative frequencies are , so the th and th values are both and the median is .
How do I compare the spread of two data sets?
Compare their standard deviations, or their interquartile ranges, and say which data set is more consistent in context. The set with the smaller value is less spread out, and it helps to compare the means or medians as well.
How do I compare two box plots?
Compare the medians for the typical value, and the interquartile ranges or ranges for the spread, and comment on symmetry. State each comparison in context, for example which group has the higher median.
How does adding a constant to every value change the mean and median?
Both increase by that constant. Adding to every value of changes the mean from to and the median from to .
Ready to practise properly?
Practice questions on every section, marked instantly like the real exam.