Research Methods: Developing your research design

Eira Patterson | View as single page | Feedback/Impact

Understanding factors and variables

This section aims to help you think about the factors that are interacting within the area you want to research. Within quantitative research these factors are called variables, which are categorised depending on the way they function in the context being studied. Exploring ways in which different types of variable are categorised in quantitative research provides insight also into the factors that need to be considered when planning a qualitative research design.

Some variables considered in quantitative research

Independent variable

This is one of the variables specified in the central research question or purpose statement or hypothesis. To identify this variable look for the one influencing one or more of the other variables in the central question or purpose statement. The independent variable, as its name suggests, is not influenced by any of the other variables in the context being studied.

Dependent variable

This is the variable which is influenced by the other variables in the system and usually the aim of the research is to study the impact of the other variables on the dependent variable. Again this variable will be specified in the central question or purpose statement.

Mediating (intervening) variables

These variables ‘transmit (or mediate) the effects of the independent variable on the dependent variable’ (Cresswell, 2011: 118). These can be grouped into two

categories:

  • mediating variables which are associated with the independent variable and it is the influence or impact of these variables which is of interest and which the research aims to explore or measure

  • mediating variables which are associated with the sample or context of the study and which often make it difficult to establish the impact of the variables associated with the independent variable

Confounding (spurious) variables

These variables ‘are attributes or characteristics that the researcher cannot directly measure because their effects cannot be easily separated from those of other variables, even though they may influence the relation between the independent and the dependent variable’ (Cresswell, 2011: 119).