Education Services homepage

Concepts and definitions 
Click on the triangles to open a section. The table below is a list of the concepts covered in each section.
Statistics
  Statistics are numerical data that have been organised to serve a useful purpose. A major role of the ABS is to provide the Australian community with statistics that will help them make informed decisions. Statistical information provided by the ABS is used widely in Australia by governments, business people, researchers, members of the public, teachers and students.
Data
Data are observations or facts which, when collected, organised and evaluated, become information or knowledge.
Data item
A data item is the smallest piece of information that can be obtained from a survey or census.
Dataset
A dataset is data collected for a particular study. A dataset represents a collection of elements; and for each element, information on one or more characteristics is included.
Outliers
An outlier is an extreme value of the data. It is an observation value that is significantly different from the rest of the data. There may be more than one outlier in a set of data.
Sometimes, outliers are significant pieces of data and should not be ignored. In other instances, they occur as a result of an error or misinformation and should be ignored. The decision to include or exclude an outlier needs to be clearly justified when discussing results.
Example:
The weights (in kilograms) of 30 students were measured and recorded in the stem and leaf plot shown in Figure 1. In this case, the stem is the whole number values and the leaves are the decimal values. The outliers are 56.3 and 67.7.  
Stem  Leaf 

56  3 
57  
58  4 4 9 
59  0 0 2 3 8 
60  0 2 4 5 7 8 9 
61  1 2 4 4 5 6 7 9 9 
62  1 2 3 7 
63  
64  
65  7 

Fig 1 Stem and leaf plot 
Variables
Sampling
Frequency and distribution
  The frequency (f) of a particular observation is the number of times the observation occurs in that data.
Cumulative frequency
Cumulative frequency is the total of a frequency and all frequencies below it in a frequency distribution. It is the running total of frequencies.
Relative frequency
Relative frequency is another term for proportion. It is the number of times a particular observations occurs divided by the total number of observations.
Distribution
The distribution of a variable is the pattern of values of the observations.  

Graphs and displays
  Graph
A graph is a diagram representing a system of connections or interrelations among two or more variables by a number of distinctive dots, lines, bars, etc.
Chart
A chart is a visual representation of data. Bar, line, pie and other types of charts are examples of charts.
Box and whisker plots (often called ‘box plots’) can be used to show the interquartile range. Figure 1 shows a box and whisker plot of student ages.
Notice that a scale is drawn underneath. Box plots can be drawn horizontally or vertically.
Frequency distribution tables can be used for nominal and numeric variables.
Example:
Twenty people were asked how many cars were registered to their households. The results were recorded as follows: 1, 2, 1, 0, 3, 4, 0, 1, 1, 1, 2, 2, 3, 2, 3, 2, 1, 4, 0, 0. This data can be presented in a frequency distribution table – see Figure 2.
Stem and leaf plots are a convenient way to organise data. Each observation value is considered to consist of two parts  a stem and a leaf.
 the stem is the first digit or digits
 the leaf is the final digit
Example:
The number of books ten students read in one year were as follows: 12, 23, 19, 6, 10, 7, 15, 25, 21, 12.
In ascending order, these are: 6, 7, 10, 12, 12, 15, 19, 21, 23, 25. Figure 3 is a stem and leaf plot of this data.
In the stem and leaf plot (fig 3):
 the stem '0' represents the class interval 09
 the stem '1' represents the class interval 1019
 the stem '2' represents the class interval 2029.
If there are a large number of observations for each stem, the stem can be split in two. For example the interval 09 could be split into intervals 04 and 59. The stem would then be written as 0(0) and 0(5).
Time series
A time series is a collection of observations of welldefined data items obtained through repeated measurements over time. For example, measuring the value of retail sales each month of the year would comprise a time series.
Trend
The ABS defines a trend as the long term movement in a time series without calendar related and irregular effects, and is a reflection of the underlying change in that measure. It is the result of influences such as population growth, price inflation and general economic changes.  
Fig 1 Box and whisker plot
Number of cars (x)  Tally  Frequency (f) 

0  l l l l  4 
1  l l l l l
 6 
2  l l l l
 5 
3  l l l  3 
4  l l  2 

Fig 2 Frequency distribution table
Stem  Leaf 

0
1
2  6 7
0 2 2 5 9
1 3 5 

Fig 3 Stem and leaf plot 
Summary statistics
  Mean
The mean of a numeric variable is calculated by adding together the values of all observations in a dataset and then dividing by the number of observations in the set. It is often referred to as the average. Thus:
Mean = sum of all the observations ÷ number of observations
For example, find the mean of these numbers 5, 3, 4, 5, 7, 6.
Mean  = (5 + 3 + 4 + 5 + 7 + 6) ÷ 6 
 = 30 ÷ 6 
 = 5 
Notice that the value of every member of the dataset is used to calculate the mean.
Median
The median is the middle value of a set of odd numbered data, or the mean of the middle two in an even numbered set after the data have been placed in ascending order.
For example, dataset A contains 3, 7, 1, 9, 2, 5, 9. Rearranged in ascending order it becomes: 1, 2, 3, 5, 7, 9, 9. The middle number is 5 so, the median is 5.
Dataset B contains 1, 3, 4, 5, 10, 12, 13, and also has a median of 5 although the values of the data vary considerably.
The position of the median can also be found by using the formula (n + 1) ÷ 2 , where n is the number of values in a set of ordered data.
For dataset A: n = 7
So the position of the median  = (7 + 1) ÷ 2 
 = 8 ÷ 2 
 = 4 
The median is the fourth number which has a value of 5.
The above example is for an odd number of observation, i.e. n = 7. However, an extra step is necessary when the number of observations is even.
For example, if n = 8 then
the position of the median  = (8 + 1) ÷ 2
= 9 ÷ 2
= 4.5 
This means that the position of the median lies between the fourth and fifth observations. To find the value of the median, add together the fourth and fifth observations and divide by two. For example, if the dataset is: 1, 1, 4, 4, 8, 9, 9, 10 then the median is, (4 + 8) / 2 = 6.
The median value is decided by its location in the ordered dataset and not because of its actual value. Notice that the values of the other members of the dataset are not taken into consideration, only their position. There are as many values above the median as there are below.
The median is usually calculated for numeric variables but may also be calculated for an ordinal nominal variable.
Mode
The mode is the most frequently observed value in a dataset. Mode is the only measure you can use when the data is categorical and has no order – for example, place of birth, favourite colour and hair colour. As the dataset is not numbers, you cannot add and divide, so you cannot find a mean. The dataset cannot be sorted from smallest to largest so you cannot find the middle value and median. The mode does not necessarily give an indication of a dataset’s centre. A set of data can have more than one mode (see Figure 1).
For example, a group friends in Year 10 have the following hair colours: red, brown, blonde, black, blonde, black, brown, brown, black, blonde, brown, brown, black.
HAIR COLOUR  FREQUENCY 

Red 1  1 
Brown 5  5 
Black 4  4 
Blonde 3  3 

The most common hair colour is brown so the mode is brown.
Range
The range is the actual spread of data including any outliers. It is the difference between the highest and lowest observation.
Range = maximum value – minimum value
For the following dataset of students' ages: 17, 15, 14, 16, 14, 15, 16, 12, 17, 13, 12, 17, 13, 16, 15
Maximum value
Minimum value  = 17
= 12 
Range  = maximum value – minimum value 
 = 17 – 12 
 = 5 
The range of the student's ages is 5 years.
Quartiles
Quartiles divide data into four equal groups. Using the example of 15 students above, we have the following ordered dataset: 12, 12, 13, 13, 14, 14, 15, 15, 15, 16, 16, 16, 17, 17, 17. We can divide this set into four equal sized groups with each group containing one quarter of the data:
 The first quartile (Q1) is the value that 25% of the data is below.
 The second quartile (Q2) is the value that 50% of the data is below. This is the same as the median.
 The third quartile (Q3) is the value that 75% of the data is below.
In the example:  Q1 = 13
Q2 = 15
Q3 = 16 
Interquartile range
The interquartile range refers to the middle 50% of data. Another way to put it is the interquartile range is the difference between the upper (75%) and lower (25%). The interquartile range is an indicator of the spread of the data. It eliminates the influence of outliers since the highest and lowest quarters are removed. The interquartile range is found by subtracting Q1 from Q3.
Five number summary (quartiles)
This is a useful way to summarise data. It consists of:
 the lowest value
 the highest value
 the first quartile (Q1)
 the third quartile (Q3)
 the second quartile (Q2).
The range can be found from the difference between the highest and lowest value. The median is the second quartile (Q2) and the interquartile range is the difference between the third and first quartiles (Q3 – Q1).
Standard deviation
Standard deviation (s) is the measure of spread most commonly used when the mean is the measure of centre. Standard deviation is most useful for symmetric distributions with no outliers.
The standard deviation for a discrete variable made up of n observations is the positive square root of the variance as show in Figure 3.
 
Fig 1 Unimodal, bimodal and multimodal
Fig 2 Quartiles
Fig 3 Standard deviation formula

