A sample refers to the group of people chosen from the population being studied that represent that population.
Variables refers to any element in research that can be measured.
To access further information on this page, click on the quick links in the box below.
Variables, elements that can be measured, are important in research as they can have numbers attached to them, allowing the researcher to formulate valid conclusions. Variables need to be identified because they show the basis upon which research participants make judgements or form their opinions about something.
Types of variables in studies showing causal relationships:
Change variables: those that are able to bring about change in the area of study. These are called independent variables
Outcome variables: affect or influence the link between cause and effect. These are called dependent variables
Linking variables: complete the relationship between cause and effect variables.
Other types of variables:
Independent variables - the cause of change in a situation
Dependent variables - the change occurs when an independent variable is introduced to the situation.
Extraneous variables - factors operating in a real-life situation that, while not part of the study, may affect the independent or dependent variables.
Intervening variable - this links the independent and dependent variable. This variable is sometimes required to establish the relationship between the independent and dependent variables (Kumar, 2014).
For example:
(Kumar, 2014, p. 88).
•reduces sample bias
•leads to higher quality data collection: the sample represents the wider population.
•reduces skewed results: when the population is vast and diverse, it is essential to have adequate representation.
•creates an accurate sample: helps the researchers plan and create an accurate sample. This helps to obtain well-defined data
Types of probability sampling
Simple random sampling: respondents chosen by chance.
Cluster sampling: population is divided into clusters or groups that represent the population. Choice is based on demographics such as age, gender, location…
Systematic sampling: sample members are chosen at a given point in time. This is then repeated at certain intervals.
Stratified random sampling: population is divided into smaller groups that do not overlap but represent the entire population.
(Kumar, 2014).
•Sample selected by the researcher rather than on a set of fixed criteria.
•Can lead to skewed results
•May not be representative of the wider population.
•Can be useful for gathering preliminary data or to reduce research costs.
Types of non-probability sampling
Convenience sampling: based on ease of access to subjects e.g. passers by, people at the Plaza, rather than subjects being representative of the population.
Judgmental or purposive sampling: chosen at the discretion of the researcher based on the purpose of the study and the knowledge of the subjects.
Snowball sampling: used where the subjects are hard to track down or the topic is sensitive e.g. homeless people or those with HIV Aids.
Quota sampling: subjects chosen on a pre-set standard.
Expert sampling: participants are known experts in the field.
Accidental sampling: data collection is finished when the desired number of responses has been received.
(Kumar, 2014).
The SAGE handbook of applied social research methods by 
Essentials of business research methods by 
Research methods and statistics for cross-cutting research: Handbook for multidisciplinary research by