Research paradigm

Phenomenography emerged in the mid-1970s from the landmark study by Marton and Saljo1 2 where they endeavoured to explain why students at the same university and course arrived at different solutions for the same problems. Phenomenography is a second-order perspective because it focuses on participants and is concerned with understanding the collective view rather than understanding an experience through multiple individual experiences. A key assumption is that individual experiences are ‘logically related’ when phenomena they experience are the same.3 Ontologically, a researcher assumes there is more than one way of experiencing the world; furthermore, the internal and external worlds are related through an individual’s awareness of the world.4 That is, if a phenomenon is outside the awareness or experience of an individual, then they will be unaware of its existence. The key epistemological assumption is that because individuals experience a phenomenon in different ways, their awareness and description of that phenomenon will be different but related and are likely to change over time.4 When undertaking a phenomenographic study, researchers are concerned with uncovering the structure and characteristics of the experience as well as the inter-relationships between the similarities and the differences.3 5 This three-dimensional understanding of the phenomenon seeks to explore both ‘what’ is experienced and ‘why’ there are similarities and differences in this experience among study participants. Therefore, since the aim of phenomenography is to uncover the collective lived experience in all its complexity and variability, the method is ideally suited for the difficult multifaceted problems encountered by nurses and midwives.

Research design

Research design in a phenomenographic study that follows methods commonly used in qualitative research; however, the researcher needs to develop an approach that uncovers the conceptions of how participants experienced the phenomena as a collective, not as individuals.6 Purposive sampling is used to recruit participants, although a larger sample size may be needed so that both similarities and variations in perceptions are gathered.4 Additionally, researchers may need some awareness of these variations so that they can sample individuals who may have experienced these.6 For example, if a diabetic educator was seeking to understand the acceptability of an education programme, they would need to include individuals who completed the programme as well as those whose attendance was incomplete. Data collection needs to ensure that the participants’ conceptions of the experience are collected so that the researcher can vicariously experience the phenomenon. Most commonly this is likely to be a semi-structured interview; however, adding observation and think-aloud data collection provides multiple sources of data for triangulation.4 7 When using interviews, a researcher needs to use a dialogic approach that encourages participants to describe and reflect on their experiences and listen empathetically, so that they are immersed in the participant’s experience and can hear the meanings and interpretations developed by the participant.4 7 Lastly, researchers may need to undertake a bracketing process, so that they are able to set aside their own conceptions of an experience and truly immerse themselves in the participants’ experience.4

The analytical approach in phenomenography focuses on developing an understanding of the qualitatively different ways individuals experience a phenomenon. This analysis seeks to develop a picture of the phenomenon from the outside (the external horizon) so that it is seen separately to the environment in which it occurs and the inside (internal horizon) such that the internal relationships and variation in experiences are revealed.4 7 Akerlind3 discusses important principles. Data analysis is an iterative process and researchers need to set aside predetermined views and refrain from making judgements too soon. Researchers need to focus on the collective data and avoid focusing on individual experiences.4 Lastly, search for meaning across this collective data to identify the variation in experience and the structural relationships.3

Data analysis processes used in phenomenography are evolving, given that it is a relatively recent research method.6 During this process, the researchers need to enter the world of their participants to vicariously experience the phenomenon. There are two steps in common with other qualitative methods, namely becoming intimately familiar with the data and that the process is cyclical, with researchers moving repeatedly between the preliminary findings and data. While there are several processes described in the literature, a seven-stage process8 has been used to examine a broad range of experiences.9–13

Research outputs

The research results in phenomenography are expressed as an ‘outcome space’ because of the need to show the qualitatively different ways the phenomenon has been experienced. The basic unit of description in phenomenography is a ‘conception’ which could be a way of experiencing or understanding. A conception has both a referential aspect, or meaning of the experience, and a structural aspect where the combination of features illustrates the variation in experience.14 Typically, researchers will present their findings as a diagram, figure or a relational table because of the need to explain how the experience looks from both the inside (internal horizon) and the outside (external horizon). The internal horizon includes categories and features to show the similarities and differences in participant experience. This outcome space could be structured hierarchically, temporally or by explanatory power to show how the various categories are related (share features) but different (how these features vary).7

Examples of use in nursing and midwifery research

Phenomenography has been used to examine a broad range of phenomena with studies providing a comprehensive and nuanced understanding of complex experiences. Berg and colleagues15 examined postoperative recovery following day surgery and identified three different experiences: conditions for recovery at home, the rollback to ordinary life and being a cog in a flow of care. These findings emphasise the need for an individualised approach to help individuals prepare for and recover from day surgery. Marynowski-Traczyk and colleagues13 found that emergency department nurses held complex conceptions of recovery for individuals with mental health conditions ranging from ‘recovery does not occur’ through to ‘ED being a conduit to recovery by connecting individuals to other services’. The authors conclude that the disparity between how emergency and mental health services understand recovery has significant implications for the care provided to this vulnerable group.

Ethics approval

Not applicable.