4.3 · SLPaper 1Paper 2

Central tendency and dispersion

Builds on 4.2 Data presentation28 min read

In a minute

The , and describe the centre of a data set. The and describe its spread. You find them for lists and frequency tables, estimate the mean of grouped data, and see what shifting or scaling the data does.

You'll be able to

  • Find the mean, median and mode of a list or a frequency table, by formula and with your GDC.
  • Estimate the mean of grouped data from , and find the .
  • Find quartiles and the IQR by hand and with your GDC, and explain why the two can differ.
  • Find the standard deviation and with your GDC, and say what they measure.
  • Say how adding to or multiplying every value changes the mean and standard deviation.
Before you start3 quick questions on what this builds on
Check yourself
How many of these 2020 students own fewer than 22 pets? pets0123f5843\begin{array}{c|cccc} \text{pets} & 0 & 1 & 2 & 3 \\ \hline f & 5 & 8 & 4 & 3 \end{array}
Check yourself
Masses, mm grams, are grouped as 40≤m<5040 \le m < 50, 50≤m<6050 \le m < 60 and 60≤m<7060 \le m < 70. Which class holds a mass of exactly 6060 g?
Check yourself
A box and whisker diagram has minimum 88, Q1=14Q_{1} = 14, median 1919, Q3=25Q_{3} = 25 and maximum 4040. What percentage of the data lies between 1414 and 2525?

Revise first: 4.2 Data presentation

Mean, median and mode

Nine students score 2,5,1,4,3,4,0,4,42, 5, 1, 4, 3, 4, 0, 4, 4 on a 55-question quiz. What is a typical score? Three different averages answer this.

Worked exampleFind the mean, median and mode of the quiz scores 2,5,1,4,3,4,0,4,42, 5, 1, 4, 3, 4, 0, 4, 4.

Try it first, then reveal one step at a time.

  1. Mean: the total is 2727 and n=9n = 9, so xˉ=279=3\bar{x} = \frac{27}{9} = 3.

In a frequency table, each value xix_{i} appears fif_{i} times, so it adds fixif_{i}x_{i} to the total. This formula is in the formula booklet.

In the formula booklet
xˉ=∑i=1kfixin,where n=∑i=1kfi\bar{x} = \frac{\sum_{i=1}^{k} f_{i}x_{i}}{n}, \quad \text{where } n = \sum_{i=1}^{k} f_{i}

Here kk is the number of different values. The total frequency nn counts data values, not columns.

Worked exampleThe number of people, xx, in each of 3030 cars at a checkpoint is shown. x12345f149421\begin{array}{c|ccccc} x & 1 & 2 & 3 & 4 & 5 \\ \hline f & 14 & 9 & 4 & 2 & 1 \end{array} Find the mean, median and mode of xx.

Try it first, then reveal one step at a time.

  1. Mean: ∑fixi=14+18+12+8+5=57\sum f_{i}x_{i} = 14 + 18 + 12 + 8 + 5 = 57, so xˉ=5730=1.9\bar{x} = \frac{57}{30} = 1.9.

On Paper 2, your GDC finds the mean from a list of values and a list of frequencies (keystrokes in the next section).

Check yourself
Find the median of 12,7,15,9,10,1812, 7, 15, 9, 10, 18.
Check yourself
Find the mean of xx. x0123f4763\begin{array}{c|cccc} x & 0 & 1 & 2 & 3 \\ \hline f & 4 & 7 & 6 & 3 \end{array}
Check yourself
The mean of 88 numbers is 66. A ninth number is added, and the mean of all nine is 77. Find the ninth number.

Grouped data

Sometimes you only know which class each value is in. Then you can't find the exact , but you can estimate it.

The idea: treat every value in a class as if it sat at the middle of the class.

Use the booklet formula with each xix_{i} a mid-interval value. The result is an estimate, not the exact mean.

Worked exampleThe journey times, tt minutes, of 4040 commuters are shown. tf0≤t<10610≤t<201520≤t<301130≤t<408\begin{array}{c|c} t & f \\ \hline 0 \le t < 10 & 6 \\ 10 \le t < 20 & 15 \\ 20 \le t < 30 & 11 \\ 30 \le t < 40 & 8 \end{array} (a) Write down the modal class. (b) Find an estimate of the mean journey time.

Try it first, then reveal one step at a time.

  1. (a) The classes have equal widths, and the highest frequency is 1515. So the modal class is 10≤t<2010 \le t < 20.
Worked exampleThe masses, mm grams, of 6060 apples are shown. mf80≤m<1005100≤m<12013120≤m<14021140≤m<16014160≤m<1807\begin{array}{c|c} m & f \\ \hline 80 \le m < 100 & 5 \\ 100 \le m < 120 & 13 \\ 120 \le m < 140 & 21 \\ 140 \le m < 160 & 14 \\ 160 \le m < 180 & 7 \end{array} Use your calculator to find an estimate of the mean mass. Give your answer to 3 s.f.

Try it first, then reveal one step at a time.

  1. Put the mid-interval values 90,110,130,150,17090, 110, 130, 150, 170 into one list, with the frequencies 5,13,21,14,75, 13, 21, 14, 7 beside them. Write both lists in your answer: that earns the method mark.
Check yourself
Find an estimate of the mean of xx. xf0≤x<434≤x<858≤x<122\begin{array}{c|c} x & f \\ \hline 0 \le x < 4 & 3 \\ 4 \le x < 8 & 5 \\ 8 \le x < 12 & 2 \end{array}
Check yourself
What is the modal class of these seedling heights, hh cm? hf10≤h<20720≤h<301230≤h<401540≤h<506\begin{array}{c|c} h & f \\ \hline 10 \le h < 20 & 7 \\ 20 \le h < 30 & 12 \\ 30 \le h < 40 & 15 \\ 40 \le h < 50 & 6 \end{array}

Range, quartiles and the IQR

A measure of says how spread out the data are.

The simplest is the range, largest value minus smallest. One unusual value can change it a lot.

1. To find the quartiles by hand, order the data and find the .

2. Split the data into a lower half and an upper half. When nn is odd, leave the median out of both halves.

3. Q1Q_{1} is the median of the lower half, and Q3Q_{3} is the median of the upper half.

The (IQR) is the spread of the middle half of the data, so extreme values don't affect it. It is in the booklet.

In the formula booklet
IQR=Q3−Q1\mathrm{IQR} = Q_{3} - Q_{1}
Worked exampleThe ages of the 1111 members of a choir are 34,19,52,41,27,63,45,38,29,56,4734, 19, 52, 41, 27, 63, 45, 38, 29, 56, 47. Find Q1Q_{1}, Q3Q_{3} and the interquartile range.

Try it first, then reveal one step at a time.

  1. In order: 19,27,29,34,38,41,45,47,52,56,6319, 27, 29, 34, 38, 41, 45, 47, 52, 56, 63. The median is the 66th value, 4141; leave it out of both halves.

Other methods exist: some keep the median in both halves, and spreadsheets often interpolate. So the same data can give slightly different quartiles.

Your GDC uses one fixed method. In their usual settings, the three models in these notes leave the median out when nn is odd, as above. A by-hand answer can differ from the GDC's, and each is right for its method.

Worked exampleThe numbers of goals, xx, in 4040 football matches are shown. x012345f7119742\begin{array}{c|cccccc} x & 0 & 1 & 2 & 3 & 4 & 5 \\ \hline f & 7 & 11 & 9 & 7 & 4 & 2 \end{array} (a) Use your calculator to find Q1Q_{1} and Q3Q_{3}. (b) Show that the data have no outliers.

Try it first, then reveal one step at a time.

  1. (a) Enter 00 to 55 and the frequencies as in the apple example, and run one-variable statistics. Write down Q1=1Q_{1} = 1 and Q3=3Q_{3} = 3.
Check yourself
Find the interquartile range of 4,7,9,10,12,15,17,204, 7, 9, 10, 12, 15, 17, 20.
Check yourself
For the data 2,4,5,7,112, 4, 5, 7, 11, a student keeps the median in both halves and gets Q1=4Q_{1} = 4. Her GDC gives Q1=3Q_{1} = 3. Which statement is correct?

Standard deviation and variance

The uses only the . The uses every value: it measures how far the values typically lie from the .

Two dot plots on the same scale from 4 to 16: set A clustered near its mean of 10, set B spread out from 4 to 16 with the same mean of 10 mean 10 4 6 8 10 12 14 16 Set A 4 6 8 10 12 14 16 Set B
Both sets have mean 1010. Set B's values lie further from the mean, so its standard deviation is larger: about 4.164.16, against 1.291.29.

In the exam you find σ\sigma with your GDC only. One small case by hand shows what it means: the variance is the mean of the squared distances from the mean, σ2=∑(x−xˉ)2n\sigma^{2} = \frac{\sum (x - \bar{x})^{2}}{n}.

Worked exampleFind the variance and the standard deviation of 2,4,4,5,102, 4, 4, 5, 10 by hand.

Try it first, then reveal one step at a time.

  1. Mean: xˉ=255=5\bar{x} = \frac{25}{5} = 5.

Your GDC shows two standard deviations. In IB Maths AA you use σx\sigma_{x}, which divides by nn as above. Ignore sxs_{x}, which divides by n−1n - 1.

Worked exampleThe number of eggs, xx, in each of 4040 nests is shown. x123456f4913842\begin{array}{c|cccccc} x & 1 & 2 & 3 & 4 & 5 & 6 \\ \hline f & 4 & 9 & 13 & 8 & 4 & 2 \end{array} Use your calculator to find (a) the mean; (b) the standard deviation; (c) the variance.

Try it first, then reveal one step at a time.

  1. Enter 11 to 66 and the frequencies, run one-variable statistics and check that n=40n = 40.
Check yourself
Use your calculator to find the standard deviation of 3,5,7,93, 5, 7, 9.
Check yourself
Ana's 400400 m times have mean 58.258.2 s and standard deviation 0.60.6 s. Bea's have mean 57.957.9 s and standard deviation 1.41.4 s. Which is correct?

Adding to or scaling every value

Add 55 marks to every test score. Every score, and so the , moves up by 55. The gaps between scores don't change, so the spread doesn't either.

Double every score instead. Every score and every gap doubles, so the mean and the both double.

bb can be negative: subtracting 44 from every value lowers the mean by 44 and leaves σ\sigma unchanged. The , and mode change like the mean; the and range change like σ\sigma.

Worked exampleOne June, a town's daily maximum temperatures have mean 18.518.5 and standard deviation 3.23.2, in degrees Celsius. They are converted to degrees Fahrenheit using F=1.8C+32F = 1.8C + 32. Find (a) the mean, (b) the standard deviation and (c) the variance in degrees Fahrenheit.

Try it first, then reveal one step at a time.

  1. Every value is multiplied by a=1.8a = 1.8, then b=32b = 32 is added.
Worked exampleA data set has mean 1212 and standard deviation 2.52.5. Every value xx is changed to ax+bax + b, where a>0a > 0. The new mean is 5050 and the new standard deviation is 1010. Find aa and bb.

Try it first, then reveal one step at a time.

  1. Only aa affects the standard deviation: 2.5a=102.5a = 10, so a=4a = 4.
Check yourself
The ages of a club's members have standard deviation 7.37.3 years. Find the standard deviation of their ages in 55 years' time.
Check yourself
A data set has variance 99. Every value is multiplied by 33. Find the new variance.

Common mistakes

  • Taking one of the middle two values, or the middle column of a table, as the median. For even nn, average the middle two data values.
  • Dividing ∑fixi\sum f_{i}x_{i} by the number of columns. Divide by the total frequency nn.
  • Using class boundaries instead of mid-interval values, or leaving the frequency list off on the GDC. Check that the GDC shows the right nn.
  • Quoting sxs_{x} from the GDC as the standard deviation. In IB Maths AA you use σx\sigma_{x}.
  • Multiplying the variance by aa, or adding bb to σ\sigma. Under ax+bax + b, σ\sigma becomes aσa\sigma and the variance a2σ2a^{2}\sigma^{2}.

Exam tips

  • Write down the mode: no working needed. For Find the median or Find the mean, show the ordered data or ∑fixin\frac{\sum f_{i}x_{i}}{n} with your numbers in, for the method mark.
  • On Paper 2, write down the lists you entered before the GDC's answer.
  • Round only the final answer, to 3 s.f. unless the question asks for another precision, such as 4 s.f.

Key takeaways

In bookletxˉ=∑i=1kfixin,where n=∑i=1kfi\bar{x} = \frac{\sum_{i=1}^{k} f_{i}x_{i}}{n}, \quad \text{where } n = \sum_{i=1}^{k} f_{i}
In bookletIQR=Q3−Q1\mathrm{IQR} = Q_{3} - Q_{1}
Learn thisMedian: the n+12\frac{n+1}{2}th value in order; for even nn, the mean of the middle two. Mode: the most frequent value.
Learn thisGrouped data: estimate the mean with mid-interval values a+b2\frac{a+b}{2}. The modal class has the highest frequency.
Learn thisQuartiles: medians of the lower and upper halves, median left out when nn is odd. Other methods may differ slightly.
Learn thisStandard deviation: GDC only, using σx\sigma_{x}, not sxs_{x}. Variance =σ2= \sigma^{2}: square the unrounded value.
Learn thisx→ax+bx \to ax + b with a>0a > 0: mean →axˉ+b\to a\bar{x} + b, σ→aσ\sigma \to a\sigma, variance →a2σ2\to a^{2}\sigma^{2}.

Practise this

9 past-paper questions on 4.3 in the AA HL bank, each with a mark scheme and a worked explanation.

Practise 4.3 →