DATA COLLECTION

Statistics is an important branch of mathematics. We use statistics to collect and display

Data is raw information that can be collected through observations, surveys or interviews
which has been organised in a useful way. When numerical data is organised in a
meaningful way, the data is statistics. Statistics is often used to help people make informed
decisions.

In statistics, a variable is something that can be measured or observed and is expected to

PRIMARY DATA
Primary data is data that you collect. There are a number of ways you can collect primary
data:
• Direct observation – this allows you to focus specifically on the details or aspects that
are important to your research.
• Surveys – Considerable amounts of data can be collected, however there is an
assumption that the person completing the survey is trustworthy with their
responses.
• Interviews – although time consuming and expensive, they allow for in-depth
questions and responses. They also allow for the interpretation of facial expressions,
body language and sarcastic remarks. However, honesty may be an issue, as the
person being interviewed may feel inclined to answer how they think the interviewer
wants them to.
• Logs – can produce many valuable data regarding performances over time under
various conditions. Examples are complaint, transaction, and fault logs.

SECONDARY DATA
Secondary data is data collected from external sources such as the internet, television,
radio, newspapers, magazines, journals, research papers, reviews or hearsay. It is usually
in the form of a ready-made product, so averages, conclusions etc. are already calculated
and stated.
Using secondary data can save time and costs, but the data can lack originality as it has
been collected by other agencies. It may not be suitable or relate directly to what is being
investigated.
Primary data is more costly, but it is also likely to be more suitable and trustworthy.
Secondary data, although cheaper and relatively easy to obtain, needs to be used
cautiously.
Information is collected in different ways, depending on what it can be used for and how
accurate it needs to be. There are three main ways of collecting data:
• census
• sample
• observation

CENSUS
A census collects information from the whole population studied. The population is all the
people or objects you want data about.

In Australia, the Australian Bureau of Statistics (ABS) conducts a national census every 5
years, collecting information from every household including details of income,
employment, education, housing and many other things. A census is accurate and detailed,
but also expensive, time consuming and often impractical.

SAMPLE
A sample survey collects data from only a part of a population. It is often cheaper and
quicker than a census, but is not as detailed or accurate. Samples are common as a census
is often impractical, particularly if a population is large. The size of the sample must be
large enough for the sample to give a fair indication of the population. For the same
reason, the members of the sample should be chosen at random.

Conclusions drawn from sample surveys always involve some error. Often this error is due
to a bias in the sample or method of data collection.

• A biased sample does not truly represent the whole population.

Many statistics do not involve either a census or a sample, but instead merely observation,
collection and tabulation of numerical facts. For example:
➢ Sports
o cricket: bowling and batting averages and run rates
o tennis: first serves in, unforced errors, double faults
➢ Weather – temperatures, barometer readings, rainfall, tides
EXAMPLES:

1. State whether a census or a sample survey be used to investigate each situation.

a) The length of time an electric light globe will last.

b) The causes of car accidents in NSW.

c) The number of people who use Bright Teeth toothpaste.

2. Suggest the possible bias in each of the following samples.

a) People surveyed by phone during the day.

b) People surveyed at a train station.

c) People selected from a football crowd.

d) Ten people tested with a new drug.

3. If you wanted the local council to upgrade in swimming pools, what sample of people
would you choose to survey?
4. What would be a suitable sample to survey if you wanted to:

a) Investigate the most popular movies?

b) Collect information about the cleanliness of trains?

1. a) Sample – It would be impractical to test every light globe produced until it failed
– there would be none to sell.
b) Census – an accurate analysis of all accidents would be important.

c) Sample – it would be very time consuming and expensive to interview the whole
population to find out who uses the toothpaste.
2. a) The sample would be biased towards people who are home during the day and
have a phone. It does not include people who go to work in the daytime or do
not have a phone.
b) The sample would be biased towards people who catch the train at that station.
It does not included people who use other forms of transport or do not travel or
use a different station.
c) The sample would be biased towards people who attend football matches. For
example, there would probably be more males than females at football matches.
d) The sample would be biased by the characteristics of the ten people. A larger
sample is needed.
3. A sample of people at the swimming pool. You would expect the people who use the
pool to be biased towards this proposal.

Note: Sometimes people use biased samples to enhance their claims for selling their
product or to support a particular point of view.
4. a) People at a cinema, people hiring DVDs, cinema managers, DVD stores.

b) People who travel on trains, train cleaners.