A methodology is a framework for research that focuses on how it is possible to collect reliable and valid data about, in this instance, the social world. It’s shaped by two main considerations:
1. Our beliefs about the fundamental nature of the social world (ontological concerns).
2. How we believe is possible to construct knowledge about the world (epistemology).
In turn, these ideas shape our choice of research methods when we come to actually collect data.
Although an understanding of Realist methodology isn’t essential at a-level, if students can grasp its basics it’s very useful as a source of evaluation for alternative methodological approaches, such as positivism and interpretivism.
It also adds a slightly different dimension to arguments over whether sociology is a science.
Realism, particularly at a-level, is often portrayed as a kind of methodological hybrid, one that combines a belief in the existence of objective social structures (positivism) with the idea they are subjectively experienced and socially constructed (interpretivism). While this is, in some respects, a valid way of looking at it, realist methodology is perhaps a little more subtle and complex than this simple formulation might suggest – an idea we can explore in the following way:
1. For Realists, societies consist of social structures that can be objectively studied because these structures have an independent existence from the people who move through them. Social structures, therefore, represent ‘real forces’ that act on, shape and in some ways determine our everyday political, economic and cultural lives.
2. While the real features of social systems make it possible to establish causal relationships, Realism breaks with positivism because it adds the proviso causality will be limited in time and space: what is true in one society / social context may not be true in another. This follows because for Realists there is a further dimension to understanding human behaviour – a subjective one that recognises and takes on board the importance of human meaning and interpretation.
The significance of these ideas is encapsulated by Searle’s (1995) argument that social reality is multi-layered, initially on the basis of two types of fact:
- Brute facts are the things we experience as real. These are many and varied but an example might be something like social inequality as evidenced by statistical differences in income and wealth. Brute facts are objective features of social reality. They are true and will impact our lives regardless of any subjective interpretations or beliefs about them we may hold.
- Mental facts reflect meanings- what people understand by their experience of brute facts. Mental facts are highly variable aspects of people’s behaviour; social inequality, for example, means different things to different people.
In this respect, therefore, that which we call “social reality” has two basic layers:
These two layers are equally important and methodologically significant because in the social world meanings always underpin the interpretation of brute facts. For Realists, therefore, the objective is not simply to understand “brute facts” (positivism) nor is it to over-determine the significance of how people interpret such facts (interpretivism). Rather, it is to understand how brute and mental facts clash and combine to produce a variety of different social experiences and outcomes (an example of which, in relation to crime and deviance, might be something like Taylor, Walton and Young’s Radical Criminology and the attempt to formulate a “fully social theory of deviance”).
3. We can understand the difference between realism and other forms of sociological methodology by noting the following:
An example here might be Durkheim’s (1897) analysis of suicide. This is often – quite understandably – seen as a classic form of positivist research:
Durkheim’s research, however, has a realist twist in the sense that he went on to explain why suicide rates varied over time and across different cultures by reference to two underlying (and not directly observable) mechanisms: levels of social regulation and integration. These, in other words, were the real causes of suicide.
4. In some respects, therefore, its understandable realism is, according to Trochim (2002), sometimes described as ‘post-positivism’, partly because both approaches believe societies have objective featuresthat can be studied scientifically. For both, social structures have an existence independent of the people who create them; they are ‘real forces’ influencing behaviour and although the social and natural worlds are different, the basic principles involved in the study of each are similar.
In this respect the ‘real’ features of social systems make it possible to establish causal relationships because structures exert a force on individual behaviour, albeit one modified through individual experiences, meanings and interpretations – ideas that mark a significant break with both positivism and interpretivism.
On the one hand, while all social action is goal-orientated (directed towards an end of some kind), we can’t understand these goals without understanding the meaning they have for people. On the other hand, meanings without goals are impossible.
What unites the two for realists is how and why they combine in particular situations: their underlying connections.
5. For realists, in the study of the social world empirical evidence is desirable but not sufficient. The structures we experience ‘as real’ are themselves the product of ‘hidden mechanisms and forces’ that are not directly observable. The social world ‘as we see and experience it’ is, in other words, governed by social processes we must understand to explain the observable world.
To take another, slightly more-contemporary, example, Soothill and Grover (1995) argue ‘sex crime’, such as rape, is both real – it’s possible to empirically identify perpetrators and victims – but also socially constructed through agencies like the media who define its meaning. To understand sex crime, they argue, it’s not enough to simply study the observable features of a situation – the relationship between the perpetrator and victim, their respective social backgrounds and so forth (positivism) – nor is it sufficient to simply dismiss sex crime in terms of social reactions (interpretivism) because it is behaviour that has real consequences for those involved. To fully understand such behaviour as criminal we must also study unobservable features, such as the ideological beliefs that surround male-female sexuality in a particular culture, that contribute to the creation of a particular social act such as rape.
In other words realist explanations, while recognising these things are important, spread the net further to dig deeper into areas that might, on the face of things, seem to have no direct relevance to a sex crime, such as aforementioned cultural attitudes to gender – do we live in a sexist society, for example? – that surround and contribute to the creation of a particular social act.
Again, for point-of-reference, this is a general approach mapped-out by Taylor, Walton and Young (1973) in their formulation of “The Seven Dimensions of a ‘Fully Social Theory of Deviance’”.
Reliability and Validity
6. To generate reliable and valid knowledge the social world must be understood in its totality – in terms of social structures, social actions and, most importantly, their relational interplay. While it’s possible to study and explain particular ‘events’ (such as a crime), to validly explain why people commit crimes we have to think more widely in terms of how the interconnected parts – both structure and action – of a social system impact on each other
In this respect Realism reflects the way natural scientists understand and study the non-human world; all phenomena are connected to each other in some way because they are all governed by physical laws. When you repeatedly drop a pen and it falls to the floor, the ‘unobservable mechanism’ that explains this regularity is gravity.
7. Realists, therefore, see reliability and validity in terms of constructing both an overall (‘in depth’) view of social behaviour in different contexts (something shared with interpretivists) and, at the same time, producing specific, causal-type explanations for behaviour (something they share with positivists).
To this end they recognise that different research methods have different strengths and weaknesses; questionnaires may produce reliable data, but with low validity, while the reverse is true for covert participant observation. For realists this reflects the nature of the social world: no single method can capture its complexity. Since all have weaknesses – as Trochim (2002) argues, all research methods contain the capacity for error – the only sensible thing is to combine methods so that one type of error cancels out another. In other words, Realists tend towards a methodological triangulation that involves:
• collecting different types of data (qualitative and quantitative, primary and secondary)
• checking data reliability and validity
• comparison: different researchers using the same method can compare data for similarities and differences
• confirmation: verifying the accuracy of different types of data.
8. In this respect Realists frequently use combinations of bothquantitative and qualitative methods and data to get the fullest possible research picture. This methodological pluralism is based on the idea that while we can define qualitative and quantitative data separately, within the research process they are inseparable; quantitative data is only useful if it has meaning, while qualitative data is only useful if it has an objective.
To put it bluntly, “method doesn’t matter”; what is discovered is more important than how its discovered.
1. Social structures are experienced as real features of human collective behaviour. As such they can be studied objectively.
2. The meanings people give to their experience of social structures are similarly real and these too can be studied objectively.
3. Reliable and valid knowledge about the social world is created by understanding it in its totality – in terms of social structures, social actions and, most importantly, their relational interplay.
4. By combining research methods (quantitative and qualitative, primary and secondary) to create a form of human ethnography Realism argues we can fully explain behaviour in terms of both its structure and its action.