Quantitative research uses large samples and, as such, the findings of well-conducted studies can often be generalised to larger populations. However, it is important that studies are well-designed to avoid errors in their interpretation and/or the reporting of inaccurate results. Misleading results from quantitative studies can have serious negative implications such as wasting public money on flawed policies and subjecting service users to ineffective or harmful treatments. This chapter explores descriptive and experimental quantitative research designs and examines, through case examples, the difference between cross-sectional, longitudinal and cohort studies. Factors leading to poorly and well-constructed studies are explored, along with a discussion of the key features of well-designed randomised controlled trials, the gold-standard design for testing treatment effectiveness.

Quantitative data analysis makes sense of numerical data. We can use numbers to summarise the experiences or characteristics of a group of participants, for example their average age or the number of symptoms they report. We can also use numbers to look at people’s behaviours, experiences and views. Perhaps most importantly, we can use numbers to look at differences between groups of people or the same group over time. This can help us understand the effect of new treatment or policy initiatives, both in terms of the type of effect (e.g. does a new policy make things better, worse or leave things unchanged?) and the size of its impact (e.g. are any changes big enough to be meaningful or could they have happened just by chance?). This chapter explores some of the different approaches to analysing numerical data, examines the difference between descriptive and inferential statistics, and highlights some of the ways in which you can begin to interpret research data presented as numbers.