Andrew Jones

Challenging false dichotomies in educational research: Insights from Critical Realism

In recent years, educational research has been shaped by the rise of cognitive science, or the ‘science of learning,’ often promoted by teachers on social media, bestselling authors, government consultants and Department for Education ‘advisory groups’ (for an overview, see Perry et al., 2021; Coe & Kime, 2019; Weinstein, Madan & Sumeracki, 2018). While cognitive science offers valuable insights, its dominance risks overshadowing broader and more holistic traditions of educational research.

Many cognitive and data scientists recognise the value of combining philosophical and diverse methodological traditions that address the complexity of education as a social phenomenon (for instance, see Jørgensen, Perry & Lea, 2023). However, this perspective is often dismissed by self-styled ‘evidence-based’ educators or ‘educational myth-busters’ (you can find them easily on X, I won’t reference anyone in particular). Similarly, some vocal cognitive scientists argue that their narrow view of research is sacrosanct, dismissing alternative approaches like qualitative research (again, you’ll find them on X). These figures often overlook the benefits of integrating diverse methods or the limitations of their own paradigmatic approaches.

Positivism – the belief that knowledge stems solely from observable phenomena and measurable data – underpins a number of methodological approaches, including those studies relying solely on experiments, randomised control trials (RCTs) and/or meta-analyses, which are frequently presented as the ‘science of learning.’ By prioritising rigid metrics, these methods can marginalise alternative perspectives and dismiss the integration of qualitative research.

This blog highlights how critical realism, as developed by Roy Bhaskar, Margaret Archer and David Scott, addresses these limitations without rejecting the overall importance of cognitive science, RCTs and meta-analyses to both researchers and teachers in general. Critical realism advocates for a more comprehensive understanding of education, challenging positivism’s dominance, and exposing false dichotomies, such as the divide between quantitative and qualitative research. While Bhaskar and Archer were not solely focused on education (Archer did, however, write about education regularly), unlike Scott, their insights into the interplay between structure and agency offer a rich and integrative framework for educational research.

Bhaskar’s critical realism

Roy Bhaskar’s critical realism provides a philosophical foundation for understanding the interplay between observable and unobservable – or even unknown – occurrences in the real world. This perspective is particularly relevant in open and dynamic environments such as classrooms or playgrounds, in contrast to the controlled conditions of laboratories or the reliance on simplified variables often used in correlational studies. In such research, measurable interventions and outputs are sometimes selectively chosen and subjectively interpreted by researchers, who may assume these data points fully capture the complexities of teaching or learning. This does not mean that such data is not valuable, but it should not be considered definitive in understanding an intervention, outcome or phenomenon.

A central tenet of Bhaskar’s work is his recognition of a stratified reality, comprising three levels: the empirical (what we directly experience), the actual (events that occur), and the real (underlying mechanisms and structures). As illustrated in figure 1 below, this ontological framework supports his critique of positivism, which he argues is overly focused on the empirical level and neglects the deeper causal mechanisms driving observable phenomena (Bhaskar, 1975; 1979).

Figure 1: Bhasker’s stratified ontology (Hoddy et al, 2019)

In education, Bhaskar’s critique is particularly significant. Positivism often prioritises test scores and other measurable outcomes, typically derived from controlled experiments with adults, without addressing the systemic inequities that underpin these results. By contrast, critical realism encourages researchers to investigate deeper structures – such as socio-economic disparities, institutional biases or the personal and professional identities of participants – that shape educational outcomes.

Similarly, methods such as RCTs reflect this positivist focus. While RCTs are frequently regarded as the ‘gold standard’ in educational research, they often fail to account for the qualitative dimensions of educational practice, such as the lived experiences of students and teachers. Bhaskar’s emphasis on uncovering causal mechanisms offers a more comprehensive alternative to the reductionist tendencies of such paradigms.

Archer’s morphogenetic model

Margaret Archer builds on Bhaskar’s critical realism with her morphogenetic model, which explores the relationship between structure and agency over time. Archer critiques the positivist tendency to conflate structure and agency, either reducing human actions to structural determinants or ignoring structures altogether. Instead, and as simplified in figure 2, she offers a relational approach where structures predate agency but can be transformed through human action (Archer, 1995).

Figure 2: The morphogenetic cycle (Archer, 1995, p. 157) 

Positivism’s limitation lies in its inability to account for this dynamic interplay. For instance, positivist analyses of education policies might focus exclusively on their implementation and outcomes, ignoring the ways educators adapt or resist these policies. The RCT approach similarly prioritises outcomes over processes, neglecting the complex dynamics that shape educational practice. Essentially, Archer’s model provides a framework for understanding these processes of adaptation and resistance, demonstrating that educational phenomena cannot be fully understood through a purely positivist lens. Her work shows how false dichotomies, such as structure versus agency, are not merely theoretical problems but practical barriers to understanding and improving education.

Methodological pluralism and epistemological integration

David Scott extends the critical realist tradition into the realm of educational methodology. Unlike Bhaskar and Archer, whose broader work encompasses a range of social phenomena, Scott focuses specifically on education. He critiques the positivist emphasis on quantitative methods and the pursuit of objectivity, arguing that these approaches often ignore the subjective and contextual dimensions of education (Scott, 2010). Table 1 contextualises Scott’s critical realist perspective by categorising the key structures that influence educational practices, illustrating the interplay between individual agency, social norms and systemic forces. These may not always be apparent in positivist studies, especially where variables have already been decided, narrowed down and, therefore, bypass any other contributing or mitigating factors impacting the study.  

Table 1: Scott’s typography of structures (adapted from Stutchbury, 2022)

Type of StructureScott’s Description
EmbodiedStructures limit change. Resistance is possible but depends on individuals’ embodiment of societal norms.
DiscursiveStructures that potentially effect change. They are visible in language and social interaction.
Structures of agencyLinked to personal identity and individual reflexivity, influencing how change is perceived and enacted.
Institutional and systemicRefers to social norms, values, and roles that shape behavior within institutions and systems.
Social markersInclude gender, class or race, all of which influence the interaction between structure and agency.

Scott’s methodological pluralism directly challenges the positivist preference for measurable data, such as exam results or attendance records, over qualitative insights like teacher narratives or student experiences. Similarly, RCTs often overlook the mixed methodologies needed to fully capture the complexity of educational systems. Scott advocates for combining these methods to address their respective blind spots. Without this integration, positivist approaches risk producing incomplete and overly simplistic accounts that fail to address the deeper realities of educational systems. Scott’s focus on methodological pluralism highlights the importance of mixed methodologies in overcoming false dichotomies, such as the division between quantitative and qualitative research.

Challenging false dichotomies in educational research

Bhaskar, Archer and Scott collectively criticise the false dichotomies that persist in the social sciences, which includes education, many of which are rooted in positivist paradigms. They encourage researchers to embrace complexity and reject simplistic binaries.

Structure vs. agency

The traditional binary of structure versus agency often leads to reductionist explanations. Structural determinism reduces individual actions to systemic forces, while radical individualism overlooks structural constraints. Positivism’s focus on measurable outcomes often reinforces these reductionist tendencies. Bhaskar’s stratified ontology and Archer’s morphogenetic model demonstrate that structure and agency are mutually constitutive. For instance, analysing inequities in school funding requires understanding both systemic barriers and the ways educators and communities respond to and navigate these challenges.

Quantitative vs. qualitative methods

Scott critiques the divide between quantitative and qualitative approaches, emphasising their complementary roles. While critical realists recognise an important role for quantitative research, they argue that it must be complemented by qualitative insights to provide a fuller understanding of educational phenomena. Positivism’s preference for quantitative methods, particularly in the form of RCTs, often marginalises qualitative research, which provides depth and context. For example, while RCTs might identify statistically significant trends, they cannot uncover the nuanced, context-specific factors that influence those trends. Bhaskar’s emphasis on uncovering causal mechanisms reinforces the importance of integrating these approaches, ensuring that research captures the multifaceted nature of education.

Positivism and critical realism

As stated above, whilst many cognitive scientists understand the importance of combining both quantitative and qualitative methodologies, a far few of its adherents’ alignment with positivism often leads to reductive paradigms that prioritise empirical observation over deeper causal analysis. Bhaskar’s critical realism challenges this by bridging the gap between naive realism and radical relativism. It acknowledges the existence of independent structures while recognising that knowledge is socially mediated. This perspective enables researchers to critically examine their assumptions and methodologies, fostering more rigorous and reflective scholarship.

Practical applications in educational research

The integration of Bhaskar, Archer and Scott’s insights offers valuable tools for educational researchers:

  • Policy analysis: National education policies often reflect structural influences that shape classroom practices. Bhaskar’s stratified ontology encourages researchers to examine both the systemic origins of these policies and the ways educators adapt or resist them in practice.
  • Equity and access: Addressing educational inequities requires exploring both structural barriers and individual experiences. Archer’s morphogenetic model and Bhaskar’s focus on causal mechanisms offer a comprehensive framework for such investigations.
  • Methodological integration: Combining quantitative data (e.g., graduation rates) with qualitative insights (e.g., student interviews) provides a richer understanding of educational phenomena. Scott’s advocacy for methodological pluralism underscores the value of this approach, ensuring that both breadth and depth are considered in educational research.

Limitations of critical realism

While critical realism provides a powerful framework for addressing the limitations of positivism, it is not without its challenges. Critics argue that:

  • Complexity: The multi-layered nature of critical realism’s ontology can be difficult to operationalise in research, requiring significant methodological sophistication.
  • Accessibility: Its abstract concepts, such as stratified reality and morphogenesis, may be inaccessible to practitioners or policymakers accustomed to simpler paradigms.
  • Integration challenges: While critical realism advocates for methodological pluralism, combining diverse methods in practice can be resource-intensive and require advanced expertise.
  • The science of learning: Cognitive science, neuroscience and the so-called ‘medical model’ of educational research (see Hess, 2022) are based on scientific methodologies, which is why they are not concerned with more holistic research designs. 

Despite these limitations, critical realism remains a valuable tool for deepening our understanding of education as a dynamic and multifaceted phenomenon.

Last thoughts

By embracing the critical realist perspectives of Bhaskar, Archer and Scott, educational researchers can move beyond reductive paradigms to engage with the complexity of education as a social phenomenon. Their frameworks emphasise the interplay of structure and agency, the integration of diverse methodologies, and the critical examination of epistemological assumptions. As educational research continues to evolve, these insights offer a path towards more holistic and impactful inquiry, enriching our understanding of the dynamic and multifaceted nature of education.

References

Leave a comment