Australian Bureau of Statistics

Rate this page
Education Services
ABS @ Facebook ABS @ Twitter ABS RSS ABS Email notification service
Education Services

Education Services home page

Teacher Statistical Literacy

Back to Education Services home page

Steps In Running A Survey

Hide details for Step 1: Planning a surveyStep 1: Planning a survey

a) Identifying your question
One of the first things you need to clarify when designing a survey is exactly what you want to find out. Start by writing your question as clearly as you can. Include as much detail as possible so that everyone else will interpret the question in the same way as you.

For example, if you wanted to find out "What times do students get up in the morning?" you would need to clarify:

  • Is it a normal school day, a weekend or a holiday e.g. “What time do students get up on a normal school day?”
  • Will it matter if students have different school starting times? Compare the question “How long before school starts do students get up?”
  • What units do you want to use to collect the data? “How many minutes before school starts do students get up?”
  • How will part units be reported? Do you want data to the closest whole number? If you plan to have fractions can these be decimals?

State your definitions
You will also need to state some definitions:
  • ‘student’ is defined as Year 7 and Year 11 Australian school students
  • ‘getting up time’ is the time students get out of bed on a school day.

It is important that you maintain these definitions throughout your investigation and in any report. If your question is not clearly defined, the participants in your survey may interpret the question differently and your results won't be accurate.

b) Deciding who to include in your sample

Participant characteristics
Next you need to specify the scope of your sample. For example, are you looking at a particular age group or year level or location? In these cases. your question might be “How many minutes before school starts do Year 7 students get up compared with Year 11 students?” or "How many minutes before school starts do students in Queensland get up compared with students in South Australia?".

Sample size
Estimates are made about the total population and subgroups based on the information from the sample. Generally, larger samples will give a more accurate representation of the population. However, it can be difficult to obtain accurate information on smaller groups within the population if the sample size is small.
In addition, the level of accuracy can usually be measured. There are formulae to determine the size of the sample that should be taken depending on the level of confidence required. One of the simplest is:
Sample size = √n
(where n is the size of the population)

To allow predictions to be confidently made about the total population, samples need to be randomly selected as well as of sufficient size. For data to be selected randomly, each data item must have the same chance of being selected as any other. Pulling data items from a hat or using the random number generator on a calculator are common ways of ensuring that data are selected randomly. Data not selected randomly may be biased towards a particular outcome.

Types of Sampling
There are a number of ways that a sample can be randomly drawn from a population. For example, you may want to ensure that each subgroup of a population is represented in the same proportion as in the general population.
For more information on types of sampling see our Glossary page.

Show details for Step 2: Collecting dataStep 2: Collecting data
Hide details for Step 3: Organising dataStep 3: Organising data

After you have collected the data, it needs to be organised so that it is useful and ready to display.

Frequency tables
A useful way to record raw data is a tally table or frequency table.
A frequency table counts the number of times – or frequency – a value occurs in the data. For example, twenty people are asked "How many TVs do you have in your household?" If 2 households have 1 TV, the frequency of households with 1 TV is 2.

Frequency tables with class intervals
When a variable has a large spread, the values can be grouped together to make the data easier to manage and present.
For example, if you asked students how much time it takes them to get to school each day, their responses may vary considerably. In this case, you can group the responses together in 5 minute intervals. These intervals are called class intervals. All class intervals should have an equal range. Class intervals are usually in groups of 5, 10, 20, 50 etc.

0l l l l4
1l l2
2l l l l l6
3l l l l l l l8
Figure 1: Frequency table of number of TVs per household

1 - 5l l l l l l7
6 - 10l l l l l l l l9
11 - 15l l l l l l l l9
16 - 2011
Figure 2: Example of a frequency table showing five minute class intervals

Hide details for Step 4: Displaying informationStep 4: Displaying information

How will you present your findings? Will you use tables, graphs or both? What sort of table or graph is most appropriate?
One of the most powerful ways to communicate data is by using graphs. Data presented in a graph can be quick and easy to understand.

A graph should:

  • be simple and not too cluttered
  • show data without changing the data’s meaning
  • show any trend or differences in the data
  • be accurate in a visual sense – for example, if one value is 15 and another 30, then 30 should be twice the size of 15
Ambiguity can be reduced by
  • avoiding 3D representations
  • avoiding broken or uneven scales
It is important that you give each graph a heading. Axes must be labelled and any scale must have equal intervals. A key should be included where it is needed to interpret the data.
Different graphs are useful for different types of information and it is important that the right graph for the type of data is selected. See What graph or display to use when for more information.
Consider the type of data you have collected when choosing a display or graph.
Type of dataAppropriate display or graph
CategoricalBar graph, pie chart, dot plot
Numerical (discrete)Bar graph, histogram, line graph, box and whisker plot, stem and leaf plot, age pyramid
Numerical (continuous)Histogram, line graph, box and whisker plot, stem and leaf plot

Bar graphs – Figure 3
A bar graph is used to represent categorical or discrete numerical data. A bar graph has a gap between each bar or set of bars and the widths of the gaps and bars are consistent. The length of the bars in a bar graph is also important: the greater the length, the greater the value. A bar graph can be either horizontal or vertical – vertical bar graphs are also known as column graphs.

Horizontal bar graphs – Figure 4
The advantage of using a horizontal bar graph over a column graph is that the category labels in a horizontal bar graph can be fully displayed making the graph easier to read.

Side by side bar or column graphs – Figure 5
Side by side bar graphs are useful to compare two or more groups for the same data.

Stacked bar or column graphs – Figure 6
Stacked or segmented bar graphs are sometimes used to display a breakdown of data for a particular group – for example, the breakdown of types of internet use by sex. They are most useful when the data is shown as a percentage value and the number of categories is very small. If there are too many categories, the frequency of the datasets can be difficult to read and compare.

Histograms – Figure 7
A histogram is similar to a column graph, however, there are no gaps between columns. Histograms are used for numerical data only. Data can be either discrete or continuous and grouped or ungrouped. A histogram should have equal width columns when possible.

Line Graphs
Line graphs are used for numerical data. They show how one variable changes in relation to another. The independent variable is always displayed on the horizontal axis. It is important to use a consistent scale on each axis so you can get an accurate sense of the data.

Time Series – Figure 8
Time series graphs are line graphs that display a trend or a pattern in data over time. The slope can be upward or negative. Patterns can be seasonal (over a short period), or cyclic (over a longer period). Alternately, a time series can show a random variation.
To use a times series to make predictions, you can smooth fluctuations by using a suitable smoothing process such as deseasonalising the data.

Pie Charts – Figure 9
Pie charts, often called pie graphs, sector graphs or sector charts, show how much each sector contributes to the total. Pie charts are useful when there are only a few categories as more than five categories can make them difficult to read.
It is very important to label each sector with its value to make comparison easier.

Dot Charts – Figure 10
A dot chart can convey a lot of information in a simple, uncluttered way. Each dot can represent one or many.

Box and Whisker Plots – Figure 11
Box and whisker plots display 5 figure summary statistics: minimum, quartile 1, median, quartile 3 and maximum for a set of numerical data.
Box and whisker plots are useful for looking at the shape of a data set. In particular, parallel box and whisker plots are used to compare two data sets and are drawn on the same number line. To help identify possible outliers, 'fences' can be drawn at 1.5 x, the IQR above Q3 and below Q1. For more information, see summarising Data, Measures of Spread.

Stem and Leaf Plots – Figure 12
Stem and leaf plots are an efficient way of recording numerical data because numbers in the stem apply to all values in the leaves. They are also very useful to show the shape of a distribution and, since they order data, can be used to identify median and quartiles.
Back to back stem and leaf plots are used to compare the distribution of two data sets.

Age Pyramids – Figure 13
Age pyramids are used to represent a population age structure. Age pyramids are a very effective way of showing change in a country’s age structure over time or for comparing different countries. Estimates and projections of Australia's population from 1971 to 2050 are available on the ABS Animated Age Pyramid page.

Figure 3: Bar graph showing importance of conserving water.
Input from 0 (Not Important) to 999 (Very Important)
Source: 2010 C@S National summary table 28.

Figure 4: A horizontal bar graph

Figure 5: A side by side column graph

Figure 6: A stacked bar graph

Figure 7: A histogram showing data with grouped intervals

Figure 8: A time series graph

Figure 9: A pie chart

Figure 10: A dot plot with many to one correspondence

Figure 11: Parallel box and whisker plots

Figure 12: A back to back stem and leaf plot of arm span

Figure 13: An age pyramid (Source: Australian Bureau of Statistics Education Services)

Show details for Step 5: Analysing the dataStep 5: Analysing the data
Show details for Step 6: Drawing conclusionsStep 6: Drawing conclusions

Commonwealth of Australia 2008

Unless otherwise noted, content on this website is licensed under a Creative Commons Attribution 2.5 Australia Licence together with any terms, conditions and exclusions as set out in the website Copyright notice. For permission to do anything beyond the scope of this licence and copyright terms contact us.