Qualitative research can involve a broad range of types of data. These can include observations of interventions, interviews, focus groups, feedback forms, documents such as diary entries or letters, and audio-visual material such as photographs, video diaries, and film. Often, qualitative research draws on multiple sources of data, with information then pooled to make sense of the bigger picture.
There are also specific analytical techniques that can be used depending on the study approach. Many techniques use thematic analysis, in which patterns or ‘themes’ that emerge from the data are identified. Sometimes, these themes are allowed to emerge organically from the data, whereas other times searches are made for specific themes. For example, if a study is looking to see whether participation in an intervention leads to specific outcomes such as enhanced wellbeing, a wellbeing model such as the PERMA model (which looks at five indicators of wellbeing: positive emotion, engagement, positive relationships, meaning, and accomplishments (9)) may be applied to see whether there is evidence that any of these specific indicators are affected. However, as well as thematic analysis, each different qualitative approach outlined above also has its own set of analytical techniques. For example, in case studies and ethnographies, detailed descriptions of the setting or individuals are often presented, followed by discussions around these descriptions. In narrative research, structural devices such as plots are constructed to tell the story. In phenomenological research, significant statements are coded into units of meaning. And in grounded theory, information is coded and positioned within a theoretical model. Where data are coded or themes identified, more than one researcher may be involved so that codes can be cross-checked.
As qualitative data can be time-consuming to collect and also yield a rich body of data, often not all participants involved in an intervention are involved in qualitative research, but instead just a sample are selected. This selection or sampling can either be representative (i.e. it aims to reflect a larger entity, such as the overall demographics of the intervention group) or non-representative. Representative samples can either be random (where participants have an equal chance of being selected to be interviewed or be involved in observation) or stratified (in which researchers split participants into different demographic groups such as ages or ethnicities and then sample randomly from within these groups). Non-representative samples can involve quota sampling (similar to stratified sampling but with participants volunteering rather than being randomly selected from within different demographic groups), purposive sampling (whereby specific participants are invited who have a particular relevance for the research question), or convenience sampling (where participants are asked to volunteer to be involved).
For a clear and readable introduction to qualitative research approaches and methods, Qualitative Research edited by David Silverman (2016) gives an overview, providing a useful framework for more in-depth reading.(10)