Quantitative research approaches
There are as many different ways of approaching research as there are ways of answering a question.
Last week we looked at how different research methods produce different types of data and evidence and thought about a question you might be interested in asking. Then we looked at whether a quantitative, qualitative or mixed methods approach would help to answer it.
In this step, we’ll outline some of the most common quantitative research approaches. While you read them, keep your thoughts about your problem and your method in mind to see if any of the approaches outlined (or a mix of these approaches) might help you to make an evidence-based recommendation to solve your problem.
Experimental research approaches test hypotheses about cause and effect. It happens under test conditions, which means the researcher tries to control the environment so that different factors, or variables, can be tested in isolation. An independent variable is changed (eg oxygen is removed), and the effect of the change on a dependent variable is observed (eg an open flame goes out). If it is not possible to control all the relevant variables, it can be difficult to demonstrate cause and effect.
Observational, or descriptive, research examines a naturally occurring situation to see if you can predict what will happen in similar situations. You might observe the way people make decisions about their retirement savings, for example. Observational research is likely to include a mixture of qualitative and quantitative data.
Correlational research examines the statistical relationship between two different things. Examples include the correlation between diabetes and odontogenic (tooth and gum) infections and body image and eating habits. It is especially concerned with determining the nature of the relationship. Correlations can be:
- positive (eg rain brings customers in)
- negative (eg rain keeps customers away)
- null, or there is no relationship (eg rain makes no discernable difference).
Comparative studies are used to compare two situations where we don’t have any way to control the environment. Comparing and contrasting elements of situations can help us to understand the factors that might lead to certain outcomes. For example, you might compare whether parents of teenage sons or daughters are more likely to get a divorce.
Simulations are models that are built at a small scale to predict the effect of changes to variables. For example, you could simulate how a wind turbine respond to extreme weather conditions. Simulations can be tested and adjusted by seeing if your predictions are replicated in the real world.
Take a look at the select bibliography for more detailed information about each approach to research.
Many of these methods are also described by Walliman (2011).
Choose one research approach from this list that might be applicable to your research project. Find out a little bit more about it and, in the comments, let us know what you have discovered and where we could find out more.
Walliman, N. (2011). Research methods: the basics. London; New York: Routledge, 2011.
© Deakin University and Griffith University