As Jurgenson (2014) notes, positivism reflects the idea that, “if enough data can be collected with the “right” methodology it will provide an objective and disinterested picture of reality” and it is, in this respect, based upon two fundamental beliefs about the social world:
1. It involves patterns of behaviour that are capable of being discovered through systematic observation / research.
2. It has an objective existence, governed by causal relationships, over and above the control of individual social actors.
This idea of objectivity is both a key strength – it suggests a social world that exists in a state that can be both described and explained separately from the hopes and desires of individual social actors – and weakness here: in order to systematically research an objective social world the researcher must be objective too. They must, for example, avoid participating in or influencing the behaviour being studied. This, however, has always been easier said than done, given the existence of the observer effect: the claim that any attempt to measure human behaviour changes that behaviour it in some unknown – and unknowable – way.
In other words, although there are a variety of research methods available to positivist researchers – from questionnaires through lab experiments to naturalistic observation – most involve an artificial situation in which the research is conducted, an awareness on the part of those being studied researched they are being researched or some sort of interaction, however minimal, between researcher and researched.
Or in some cases, all three.
The relatively recent development of computer networks, such as the Internet, has arguably breathed new life into positivism through the medium of Big Data: the ability to collect massive amounts of tiny fragments, almost insignificant in themselves, of information from hundreds, thousands and even millions of individuals as they participate in their everyday, online, lives on various types of social network.
These include both obvious, such as Facebook, Twitter and Instagram and less-obvious candidates: dating sites and apps, for example, where behaviour can be monitored in real time, manipulated in various ways and recorded for later analysis.
The usefulness of huge amounts of data that can be collected relatively cheaply and efficiently can be expressed in terms of “the 3Vs”:
1. Volume: The more data that can be collected, the greater the researcher’s ability to understand how people behave within social networks.
2. Velocity: Big Data doesn’t just involve the collection of massive amounts of data. It’s possible for researchers to see how relationships change and flow in real time across computer networks.
3. Variety: Big Data can involve the collection of a wide variety of quantitative data that would be impossible to collect in any other way – from how people interact individually in a network to how millions of simultaneous interactions create structured forms of interaction.
The ability to collect Big Data has two broad advantages for neo-positivists:
1. It is unobtrusive: people are unaware their behaviour is being monitored and that data is being collected. This makes it possible to observe and understand how people “really behave” in their everyday (online) lives.
2. It is objective in the sense that data isn’t being collected by a human researcher, as such, but by a machine. How people interact in any given network – the posts they like, the pictures they post, the arguments they create – is dispassionately recorded for later analysis.
This means that, after a fashion, neo-positivism goes some way towards resolving the observer effect problem by removing the overt researcher from the equation and collecting data covertly and unobtrusively by machines.
Where Big Data involves the (anonymous and anonymised) monitoring of individual behaviour in social networks, data collection is passive: it is captured automatically without the respondent necessarily being aware of the process, which has two main consequences:
1. Researchers have no direct interaction with those being researched.
2. The objectivity, reliability and validity of the data collected is potentially enhanced precisely because there is no direct interaction between researcher and researched.
To summarise these ideas, therefore, neo-positivism involves the ideas that:
1. People are studied in a perfectly natural situation.
2. They are unaware they’re being studied.
3. Researcher and researched never meet.
The ability to collect Big Data without the providers of that data being aware their behaviour is being observed helps to reveal patterns of behaviour that operate on two levels:
1. The individual: the ability to collect information linked directly (and anonymously) to individual behaviour – from the things people buy, to what they share, like and dislike on social media – and to relate each piece to every other piece in the network of online relationships is a powerful data collection technique.
In addition, the greater the number of sample points you collect about individuals – from the “Big Things” like their occupation, education, gender, sexuality, age and ethnicity, to a range of “Smaller Things” such as their political preferences, what they look like, where and how they socialise, what they like, dislike and so forth – the greater the probability of understanding how and why they behave in the ways they do.
In this respect, when it comes to individual social actors, the more you have in terms of sample points about their behaviour, the more you know about their likely behaviour.
2. The societal: this involves seeing how individuals are linked into behavioural networks. How, for example, their apparently individual choices contribute to the creation of social facts – or indeed social structures.
From a general sociological perspective one well-known problem with quantitative research methods such as questionnaires and interviews is their openness to measurement errors that affect both the reliability and validity of the data collected. As Elizabeth Loftus, for example, has classically demonstrated, how a question is worded can fatally undermine both the reliability and validity of any data captured.
By removing the researcher from the data collection process – data is captured electronically without the need for the researcher – it is data captured in its “purest form”, uninfluenced by the presence of the researcher.
A major drawback to this kind of research involves ethical questions relating to the consent of research participants. This is often a grey area in sociological research and the kinds of secretive forms of observation carried-out by some types of neo-positivist research mirror, in some ways, more-qualitative techniques such as covert participant observation.
For both it’s essential that those being researched are unaware of their participation since, in their different ways, the objective is to observe people and groups “behaving normally”.
As Jorgenson puts it: “They cannot know they are in a lab” – something that raises all kinds of ethical questions relating to participant consent, possible harms, the right to withdraw from research and so forth.
The bottom line here is that if those being researched must not be allowed to know they are the subject of sociological research, in case this “biases the data”, there is no possibility of their giving their consent to their participation.
A second and perhaps more-fundamental criticism of this type of neo-positivist research is that, as Jorgenson argues, “Big Data positivism myopically regards the data passively collected by computers to be objective. But computers don’t remember anything on their own. This naive perspective on how computers work echoes the early days of photography, when that new technology was sometimes represented as a vision that could go beyond vision, revealing truths previously impossible to capture…But at the same time early photography often encoded specific and possibly unacknowledged understandings of race, gender, and sexuality as “real.” This vision beyond vision was in fact saturated with the cultural filter that photography was said to overcome.
This leads to perhaps the most dangerous consequence of Big Data ideology: that researchers whose work touches on the impact of race, gender, and sexuality in culture refuse to recognize how they invest their own unstated and perhaps unconscious theories, their specific social standpoint, into their entire research process. This replicates their existing bias and simultaneously hides that bias to the degree their findings are regarded as objectively truthful”.
Finally, the reliance on data from online social networks means that neo-positivist analyses of this world may not have any direct relevance to our understanding of offline behaviour in the non-virtual world.
That is, there is an implicit assumption that our online behaviours directly mirror our offline behaviours. Whether or not this assumption is warranted is, at the very least, open to question…