Chapter 2 – Processing and Representing Data

Tables
Databases – Tables with a collection of data. They are a form of secondary data is the data is available
online and, in most cases, easily accessible.
These tables usually contain information from real-life statistics, and you will be asked in the exam to
extract and interpret information from it. These questions have multiple parts and many 1 marker sub-
questions. You need to be able to use these tables to
identify values, calculate totals/differences/percentages,
describe trends and explain inconsistencies. One of the
main inconsistencies will be that the percentages do not
add up to 100% and this is due to rounding errors
because individual percentages for columns/rows in the
tables have been rounded.
As the data represents real-world statistics you may be
asked to explain reasons for trends. Think about the data
in terms of real-life rather than just an exam question.
What real-life situation may affect the data you have?

Two-Way Tables – Has information in two categories and has

two variables so the data is called bivariate data.
To find missing values, start with the row or column that has only
one value missing. Make sure the grand totals for the rows and
columns add up to the same number.
When comparing data from two-way tables, write about
comparisons between rows/columns but also individual cells.

Pictograms
Uses pictures or symbols to represent a particular amount of data. Always has a key to show the amount
each symbol represents.
When drawing a pictogram, make sure that:
 Each symbol is the same size
 The symbols represent numbers that can be easily divided
to show different frequencies, e.g. for a symbol that
represents 4, you can draw a quarter of the symbol to
show a frequency of 1.
 Spacings are the same in each row.
 There is a key to show the frequency that each symbol
represents.
Bar Charts
 Simple Bar Charts
o Bars are equal width
o Equal gaps between bars
o Frequency on y-axis

 Vertical Line Graph

Similar to simple bar chart but with lines instead of bars.

 Multiple Bar Charts

Can be used to compare two or more sets of data.
Has more than one bar for each class represented by different colours which is
shown in the key.

 Composite Bar Charts

Has single bars split into different sections for each different
category. Usually used to compare different times/days/years.
The frequency of each component should be calculated by
subtracting the upper frequency of that component with the lower
frequency. Do not just read off the y-axis (unless looking at total
frequencies or the bottom component).

Stem and Leaf Diagrams

A good way of organising data without losing any of the detail – All the original data is in the diagram
but looks simple. It also shows the shape of the distribution – whether most of the data lies at the
beginning, the end or is distributed in the middle.
Each value is split into a ‘stem’ and ‘leaf’ – Stems can be more than one digit, leaves are single digits
only. No need for commas in between leaves. Leaves must be written in order from smallest to largest
– this makes it easier to find mode and median.

How to draw one:

1) Put the first digits of each piece of data in numerical order down the left hand side.
2) Go through each piece of data in turn and put the remaining digits in the correct row.
3) Re-draw the diagram, putting the pieces of data in numerical order.
4) Add a key.

 Back-to-back Stem and Leaf Diagrams

o Shows two sets of data sharing the same stem so that you can easily compare them.
o Numbers closest to the stem are smallest.
o Use two different keys for each set of data.
Pie Charts
A way of displaying data to show how something is shared or divided into categories,
Each sector shows what proportion that category represents of the total data,
Area of Pie Chart = Total Frequency
Angles add up to 360⁰.

How to draw a pie chart:

1. Total up the frequency
2. Calculate the angle for each frequency. 360/frequency
3. Calculate the angle for each category will be by multiplying your previous answer by the frequency.
4. Make sure all the angles now add up to 360.
5. Draw the pie chart.
6. Label the sectors.

Interpreting Pie Charts – Remember pie charts show proportion and not numbers.

Comparative Pie Charts

Can be used to compare two sets of data of different sizes. The areas of the two circles should be in the
same ratio as the two frequencies.
Why? Drawing two pie charts the same size can be misleading.
Area of Pie Chart = Total Frequency
So, the larger the pie chart, the greater the frequency.
To compare the total frequencies, compare the areas.

Working out radius of second pie chart:

1. Divide both areas (this gives you the area scale factor)
2. Square root answer (this gives you the scale factor for radius)
3. Multiply by radius of first pie chart.

 If pie chart B is larger than pie chart A then pie chart B has a greater frequency.
 If both pie charts then have the same angle for a sector that means that sector has a greater
frequency in pie chart B even though the proportions are the same because it has a larger area.

Population Pyramids
Shows distribution of ages in a population, in numbers or proportion/percentages.
They are used to compare two sets of data, usually genders or two geographical areas.
When comparing the data look at the shape of the distribution.
 If it looks like a pyramid with smaller bars at the top that means there is a higher proportion of
younger people in the population and less older people. This could be because if short life
expectancy (how long people live), high birth rates or high death rates.
 If the diagram looks more or less straight that means there is a similar proportion of older and
younger people in the population which could be because of lower birth/death rates or that the
life expectancy is increasing.
 An upside-down pyramid with larger bars at the top and smaller bars at the bottom shows that
the population has a larger proportion of older people compared to younger people. This could
be because of low birth/death rates, longer life expectancy or the location might be far from
the city or a coastal area where older people are retiring to.
Choropleth Maps (not Chloropeth)
Think colour by numbers.
They split a geographical area into different regions which are then shaded.
The darker the shading the higher the frequency for that area.
Each map has a key to show what the shading represents.
Interpreting:
The area of the map which is shaded darkest has the highest proportion/percentage.
Look at the key for the shading to read off percentages/numbers.

Cumulative Frequency Diagrams

Cumulative frequency is a running total of the frequencies.
To work out CF for a class interval, add all the frequency for that class interval and the CF of the previous
class interval.
Use upper bounds for x-axis when plotting points.

CF Step Polygons – Use for discrete data.

Plot the points using upper bound of class interval and join points using straight lines
by going across then up.

CF Curves – Use for grouped continuous data.

Plot points using upper bound of class interval and connect with a smooth curve.

Estimating values from CF diagrams:

 Median
o Work out median value by dividing total frequency by 2.
o Find on Y-axis
o Draw horizontal line from that value to curve/line
o Read off value from x-axis
 Interquartile Range (IQR)
o Work out 25% and 75% values
o Find on y-axis
o Draw horizontal line from that value to curve/line
o Read off values from x-axis
o Subtract them (Big one – small one)
 Estimating more than/greater than values
o Draw a vertical line from the value in the question on the axis to the curve.
o Read off corresponding y-axis value.
o Subtract from total frequency.
Histograms
Represents continuous data from grouped frequency tables.
No gaps between bars.

Equal Class Widths

x-axis = data
y-axis = frequency
Looks like bar charts without gaps.

Unequal Class Widths

Area of bar = frequency
Y-axis = Frequency Density (not frequency)
The idea is that the frequency density reflects the ‘concentration’ of things within each
range of values.

𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦
𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝐷𝑒𝑛𝑠𝑖𝑡𝑦 =
𝐶𝑙𝑎𝑠𝑠 𝑊𝑖𝑑𝑡ℎ

𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝐷𝑒𝑛𝑠𝑖𝑡𝑦 × 𝐶𝑙𝑎𝑠𝑠 𝑊𝑖𝑑𝑡ℎ = 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦

Drawing Histograms:
1. Calculate class widths for each class interval
2. Calculate frequency density for each class interval using FD = F/CW formula.
3. Draw a suitable scale on y-axis labelled frequency density.
4. Draw bars using frequency density data. (Remember the bars have no gaps in between)

Estimating frequencies from histograms:

With these questions you are using the class widths and frequency density from the histogram to work out
frequencies. Be careful when calculating class width as some intervals may not include the entire bar.
1. Find the bars that cover the range you need from the question.
2. Work out the frequency for each bar using the FD X CW = F formula.
3. Add the frequencies.

To compare histograms, they need to have the same class intervals and frequency density scales.
When comparing histograms, describe the shape of the distribution and what this shows.

The Shape of a Distribution

This is the shape formed by the diagram. It can be positive, negative or symmetrical.
Frequency Polygons
Similar to histograms with equal class widths but without the bars.
Uses mid-points of class intervals and points are plotted and then joined together with straight lines.
Common errors:
 Midpoints not used
 Joined together at the bottom
 Points not joined together with straight lines but with a curve
instead.

Misleading Diagrams
Diagrams can be misleading because of their shape of because of axes and scales.

Types of Misleading Diagrams:

 Pictograms – Same symbol and size needs to be used for all the diagrams and key needed.
 3D charts – They distort parts of the diagram making it difficult to read off values.
 Colours – Some colours may make parts of the diagram stand out more thus making it seem more
important when it may not be.
 Lines that are drawn too thick making it difficult to read information from the diagram.

Axes and Scales that can be misleading:

 Scales that do not start at zero.
 Missing values on the scales.
 Axes that are unevenly scaled.
 Axes that are not labelled.
 Not using a key.