Chapter 2: Foundations of qualitative research – paradigms, philosophical underpinnings
Darshini Ayton and Tess Tsindos
Learning outcomes
Upon completion of this chapter, you should be able to:
- Recognise and understand the four main paradigms that underpin research.
- Understand how paradigms differ between qualitative and quantitative research.
- Describe the differences between approaches in inductive and deductive research.
What is a paradigm?
All research takes place within a paradigm. A paradigm is a worldview – a framework of beliefs, values and methods. For researchers, the paradigm or worldview framing their research informs the meaning they interpret from the data. Each researcher works within their own, unique paradigm; this includes the techniques they choose for collecting and analysing data.1 There are four main research paradigms in social science (see Table 2.1.):
- positivist, or scientific, paradigm
- interpretivist, or constructivist, paradigm (also known as the naturalistic paradigm)
- radical, or critical, paradigm
- post–structuralist paradigm.2,3
These paradigms reflect the researcher’s beliefs about what is reality (ontology), knowledge (epistemology), the means to obtaining knowledge (methodology) and the values of the researcher (axiology).3 We might think of ontology as ‘what is true’ and epistemology as ‘how do we know those truths?’. The positivist paradigm is suited to quantitative research because it is grounded in the notion of cause and effect. The remaining three paradigms are suited to qualitative research because they are grounded in exploration and understanding.3
Qualitative research is embedded in the interpretivist, or constructivist paradigm. The understandings and beliefs of interpretivism or constructivism can be considered in terms of:
- Assumptions and values: The research seeks to understand what it is to be human, and the significance and meanings people ascribe to life events. It aims to identify what is important and what is evidence.4
- Researcher–participant relationship: The relationship is ‘intersubjective’ – that is, the researcher is the listener and interpreter of the data obtained from the participant. The researcher discovers the truth of a situation through thinking and analysis, rather than sensory observation. Interpretation is required.4
- Methodology (the research approach): Qualitative research includes study designs such as descriptive, phenomenology, action research, case study, grounded theory and ethnography.4 These are covered in section 2 of this textbook.
- Ontology (the nature of reality): The researcher recognises that there are multiple subjective realities, and that these are socially constructed in the interactions between research participants and between the researcher and participants.5
- Epistemology (the theory of knowledge, or how knowledge is created): Knowledge is derived from the everyday. The researcher creates meaning from the data through their thinking and analysis of the data informed by their encounters with participants.3
- Axiology (the value and ethics of the research): The research will reflect the values of the researcher who aims to present a balanced interpretation of the results.5
A good example of the interpretivist or constructivist paradigm is a study exploring physical and bodily pain. We humans each experience pain differently, due to many factors, including how we were socialised to respond to pain in our family and communities, our individual pain threshold, our past experiences of pain and the context of our current pain – what else is happening in that moment of pain. Thus, it is reasonable to say that pain is socially constructed. In health care, pain is measured on a numerical scale, but it is the person’s perception of the pain that determines the number assigned to their pain (self-report). We may therefore question whether there is an objective, scientific method for measuring pain. A common facilitator for pain relief – paracetamol – reduces pain for some people and not for others. Consider why this is the case. (It is the person’s perception of what is working to relieve their perceived pain.) Positivists do not rely on subjective experiences, only facts and a singular truth: objectivity. Constructionists and intepretivists contend that subjective and social experiences create reality, and that there are many truths.
Although it is not one of the four main paradigms, post-positivism is another paradigm that appears in the literature. Post-positivism asserts that there are multiple and competing views of science, and multiple truths. Therefore, researchers cannot be completely objective, unbiased and value-free, as the positivist paradigm asserts.4 This shift in perspective from positivism to post-positivism has led to the incorporation of qualitative methods into the post-positivist paradigm, to enable the research to explore participants’ experiences of the phenomenon under study. This paradigm is included in Table 2.1. since its basic ontology is similar to the positivist paradigm.6
Table 2.1. Research paradigms in social science
| Paradigm | Positivist/scientific | Interpretivist / constructivist |
Radical/critical | Post-structuralist |
|---|---|---|---|---|
| Assumptions and values | Objectivity, systematic and detailed observation. Seeks to explore cause and effect. Problem-solving. Stems from science and mathematics. | Seeks to understand what it is to be human, and the significance and meanings people ascribe to life events. Aims to identify what is important and what is evidence. | The desire to change the world and not to just describe it. The world is unjust and inequalities stem from the social lines of gender identity, ethnicity, class, age, sexuality etc. Action is required and it is possible to change these injustices. | No one can stand outside the traditions or discourses of their time. |
| Researcher/ participant relationship | The researcher is the ‘expert’ and is expected to be objective. The participant is the object of the research. | ‘Intersubjective’ – the researcher is the listener and interpreter of the data obtained from the participant. The researcher discovers the truth of a situation through thinking and analysis, rather than sensory observation. Requires interpretation. | The researcher takes a normative stance and has views or beliefs about what social structures are powerful and what should be done to change them. The relationship is characterised as co-research, reciprocal, participative, empowering and power-sharing. | The researcher analyses participants as subjects of discourse, in which the researcher is also embedded. |
| Methodologies and methods | Quantitative experimental, or non-experimental. Hypothesis-driven, statistical-testing, evidence-based practice. | Qualitative – interviews, grounded theory methodology, hermeneutics, phenomenological research. | Emancipatory action research, participatory research, collaborative research, critical ethnography, critical or radical hermeneutics, critical policy analysis. | Research focused on texts – written, spoken and visual. Methodologies allow for complexity and contradiction in data. Discourse analysis, feminist post-structuralism and queer research are examples. |
| Ontology | Social reality is stable and ordered, and made up of discrete and observed events. | Reality is subjective and socially constructed. | Social change must begin at the roots of social reality. | Power is always part of social practices and in the construction of different forms of knowledge. |
| Epistemology | Knowledge is derived from sensory observations by an objective researcher. Knowledge is sought so that people (health workers, policy makers, and professionals), can explain, predict, or control events. Knowledge is gained through testing an hypothesis. | Knowledge is derived from everyday observations. | Knowledge is socially constructed, communal, contextual and subjective; however, it is also rational and emancipative. | Knowledge is social in nature. |
| Axiology | The researcher is removed and distanced from the research to ensure that their beliefs and values do not influence the research or research interpretations. | The researcher is part of the research. The ‘what’ and ‘how’ of the research will reflect the values of the researcher. The researcher aims to present a balanced interpretation of the results based on their own understanding and the data. | The researcher seeks to change the world through their research by drawing on the experiences of marginalised and disempowered groups in society. The researcher values participation and power sharing. | The researcher is embedded in the same discourses as the research participants and aims to understand how power works through the constructed discourses of participants. |
In Table 2.2. an article is provided to highlight the different components of the research paradigms. Note: The aims are reproduced verbatim from the papers (word for word).
Table 2.2. Examples of paradigms within published research
| Paradigm | Positivist | Constructivist | Radical/critical | Post structuralist |
|---|---|---|---|---|
| Title | Medical masks versus N95 respirators for preventing COVID-19 among health care workers - a randomized trial7 | Experiences of home health care workers in New York City during the Coronavirus Disease 2019 Pandemic: a qualitative analysis8 | Hearing the voices of Australian healthcare workers during the COVID-19 pandemic9 | A Foucauldian discourse analysis of media reporting on the nurse-as-hero during COVID-1910 |
| CC Licence | CC BY-NC-ND 4.0 | |||
| Aim | 'To determine whether medical masks are non-inferior to N95 respirators to prevent COVID-19 in healthcare workers providing routine care.'[abstract] | 'To understand the experiences of home health care workers caring for patients in New York City during the COVID-19 pandemic.'[abstract] | 'To better understand the challenges Australian healthcare workers have faced during the COVID-19 pandemic.'[abstract] | 'To explore media reporting on the role of nurses as being consistently positioned as ‘heroes’ during COVID-19.'[abstract] |
| Epistemology | Knowledge is objectively measured. In this paper, the primary outcome was confirmed COVID-19 infection on a reverse transcriptase polymerase chain reaction test (RT-PCR). Sera from participants were also tested for IgG antibodies. These are objective scientific measurements. | Knowledge is created from the experiences and perspectives of the 33 home health care workers in New York City. | Knowledge is created in the form of advocacy and critical voices of the current state of play. | Knowledge is created through the analysis of discourse to understand social norms and assumptions which influence behaviour and expectations. |
| Methodology | Pragmatic, randomised, open-label, multicentre trial | Qualitative grounded theory study | Qualitative analysis of responses to an open letter to the Australian government, advocating for better respiratory protection for healthcare workers | Foucauldian discourse analysis of media reports |
| Ontology | COVID testing via RT-PCR was administered via nasopharyngeal swabs and was administered at baseline and the end of follow-up (10 weeks post-baseline). Other measures such as serologic evidence of infection via IgG antibodies, respiratory illness/infection, work-related absenteeism, and for those who were positive for COVID – intensive care admission, mechanical ventilation or death. Participants received a text message twice a week asking about signs and symptoms of COVID which triggered a PCR test if symptoms were present. These measures are examples of discrete and objectively measurable observations. |
The interview guides canvassed the everyday experiences of healthcare workers during COVID with questions on "1) What workers knew about COVID; 2) how COVID affected their work and 3) the challenges they experienced during COVID".(p1454) These questions create the opportunity for participants to share their stories which is an example of social construction. | The data collection was a social change activity. An open letter was written to the federal government, calling for better access to respiratory protection, infection-control guidelines and transparent reporting of healthcare worker infections. It garnered 3500+ signatures. Healthcare workers were invited to share their concerns and experiences, which led to 569 free-text contributions. These contributions are an example of a call for social change and action based on social reality. | Three Canadian newspapers were searched over three months for articles related to nurses and COVID-19, leading to 559 articles being identified. These were screened for relevance, and 50 articles were included in the analysis. The articles were analysed through a process called ‘making strange’, in an attempt to remove assumptions and read the stories as an outsider. Articles were read multiple times over weeks to identify discourses about nurses in the context of COVID-19. with particular attention given to the discourse of caring and disciplinary power. These newspaper articles are an example of how sensemaking of a social phenomenon (COVID-19) in the context of a particular role (nurse) creates hierarchies and power dynamics in society. |
| Axiology | This study had 29 healthcare facilities, and healthcare workers were randomly allocated to either medical masks or N95 respirators. A total of 1009 healthcare workers were enrolled (the calculated required sample size was 875 people). The randomisation and large participant numbers were to ensure generalisability and objective ability to detect a difference between COVID-19 infections in participants wearing a medical mask versus those wearing a respirator. This approach aims to reduce bias. | Three of the researchers were experienced in how to conduct qualitative interviews. They conducted the interviews and were guided by a semi-structured interview guide, which was developed based on prior research by the team, other studies and conversations with agency leaders. Hence, data collection was informed by subjective experiences and research literature. Data analysis involved three researchers in the coding process, through an analysis process called the constant comparative approach, which is a common approach in grounded theory analysis. This approach is systematic, with much interaction across time and between researchers. Because researchers are involved in the data analysis and interpretation of data, their values and experiences as individuals will shape the analysis process. | This study is fundamentally a call to action and is underpinned by the value of social justice. Some of the authors were actively involved in national advocacy campaigns, leading to strong engagement with healthcare workers for this open letter. The signatories demanded change due to what they perceived to be unfair and unjust circumstances for healthcare workers. The free-text responses provided specific examples of injustice, which were then themed to provide an overall narrative. | Fifty articles were analysed to determine how the discourses of ‘nurses as heroes’ and ‘nurses as carers’ and ‘nurses as sacrifice’ created a reality in which nurses were afraid of speaking up to advocate for their protection at work and hence compromised their own safety. |
Approaches in qualitative methods
Approaches are how a researcher intends to carry out their research. In qualitative research, there are two main approaches:
- Inductive: Driven by the participants and their data
An inductive approach employs a ‘whole world’ view and includes the wider social and historical context. It considers the layers that surround the individual – temporal, spatial, ideational, institutional and structural, and focuses on meanings, ideas and experiences. The inductive approach is concerned with participants’ subjective views. When examining what participants have said in an interview, the researcher searches for themes, setting aside preconceived notions.6 (Review the example in Chapter 3 of exploring seniors’ perceptions of health and loneliness. The theoretical drive of the research is inductive because it is describing and exploring the perceptions of seniors.) An inductive research approach is hypothesis–generating – this means the researchers do not have preconceived ideas of what they will find in their research and data, and hypotheses will be generated in the process of analysing the data.6
- Deductive: Driven by a pre-existing theory, framework or series of questions
A deductive approach can employ a theory or framework to guide the research, and responses are usually categorised into pre-determined labels (most often called ‘codes’). This is usually how questionnaires or structured interviews are interpreted. The pre-determined codes are based on the questions asked in interviews or focus groups. A deductive approach, particularly in the positivist paradigm, is hypothesis-testing – the researchers are looking for evidence of specific ideas, concepts and relationships in the research and data.5
All research takes place within a paradigm, consciously or subconsciously; that is, regardless of whether this is understood by the researcher. To interpret the data well, qualitative researchers must explore and acknowledge their own framework of beliefs, values and methods informing the meaning of their data. Qualitative research is embedded in the interpretivist paradigm. Four main paradigms have been explored and explained in this chapter.
References
- Donmoyer R. Paradigm. In: Given LM, ed. The SAGE Encyclopedia of Qualitative Research Methods. SAGE Publications; 2008:591-595.
- Denzin NK, Lincoln YS. The SAGE Handbook of Qualitative Research. SAGE Publications; 2006.
- Giddings LS, Grant BM. Mixed methods research for the novice researcher. Contemp Nurse. 2006;23(1):3-11. doi:10.5172/conu.2006.23.1.3
- Levers, M-JD. Philosophical paradigms, grounded theory, and perspectives on emergence. SAGE Open. 2013;3(4). doi:10.1177/2158244013517243
- Kivunja C, Kuyini AB. Understanding and applying research paradigms in educational contexts. International Journal of Higher Education. 2017;6(5):26-41. doi:10.5430/ijhe.v6n5p26
- Morse JM. The paradox of qualitative research design. Qual Health Res. 2003;13(10):1335-1336. doi:10.1177/1049732303258368
- Loeb M, Bartholomew A, Hashmi M, et al. Medical masks versus N95 respirators for preventing COVID-19 among health care workers: a randomized trial. Ann Intern Med. 2022;175(12):1629-1638. doi:10.7326/M22-1966
- Sterling MR, Tseng E, Poon A, et al. Experiences of home health care workers in New York City during the coronavirus disease 2019 pandemic: a qualitative analysis. JAMA Intern Med. 2020;180(11):1453-1459. doi:10.1001/jamainternmed.2020.3930
- Ananda-Rajah M, Veness B, Berkovic D, Parker C, Kelly G, Ayton D. Hearing the voices of Australian healthcare workers during the COVID-19 pandemic. BMJ Leader. 2021;5:31-35. doi:10.1136/leader-2020-000386
- Boulton M, Garnett A, Webster F. A Foucauldian discourse analysis of media reporting on the nurse-as-hero during COVID-19. Nurs Inq. 2022;29(3):e12471. doi:10.1111/nin.12471
