Quantitative research is used to answer questions that have numerical value answers. Quantitative research is also used to establish cause and effect relationships between variables.
Quantitative research designs
- Randomised controlled trial – considered to be the best design to establish cause and effect relationships. Key features of a RCT include a treatment arm/group and a control arm/group.
- Quasi-experimental – similar to RCTs with no randomisation.
- Cohort studies – follow a predetermined sample group to measure the incidence of outcomes. The purpose of cohort studies is to link an exposure to an outcome. Purely observational with no intervention from the researcher.
- Case control studies – the retrospective form of a cohort study. Individuals with the desired outcome are chosen, with the researcher attempting to discover the exposure that the outcome can be attributed to. Highly prone to recall bias.
- Cross sectional studies – used to determine the prevalence of an outcome within a specific group. Often conducted using surveys, cross sectional studies are common in healthcare due to being cheap and easy to conduct.
Types of data collection within quantitative research
- Pre-existing data
- Observation of behaviour
Strengths and limitations of quantitative research
- Data can be interpreted using statistical analysis
- Can establish cause and effect relationships
- Computer software available to analyse data – saves time and helps to minimise risk of human error
- Easy to replicate and generalise
- Do not reflect real life due to the high control applied.
- Reductionist – simplifying complex situations into simpler versions
- All confounding variables cannot be controlled
- Lacks breadth within data
Terminology associated with quantitative research
- Internal validity – whether the results are based on the intervention or an unknown variable.
- External validity/Generalisability – how well what is being measured can be generalised to the wider population.
- Confidence interval – usually expressed as a percentage. Represents how certain the researchers can be that the mean for the entire population would fall within the identified range.
- Hypothesis – a theory or idea that needs to be tested.
- P value – a measure of the strength of evidence against the null hypothesis. a small p value < 0.05 indicates evidence against the null hypothesis, this is then rejected and an alternative hypothesis developed.
- Independent variable – the variable manipulated by the researcher to measure its effect on the dependent variable.
- Dependent variable – what the researcher is interested in measuring in the study.
- Confounding variable – an outside influence that can affect the results of a study.