Data, Ethnography, Field, and Place.

Steph from ICU-Heart and Max from the DARE project have been chatting about ethnography and the concept of  place, and about how this relates to digital research in Trusted Research Environments. Here’s a co-authored blog about it!

***

Data is big in the UK health system. In the 2025 NHS 10 Year Plan, the UK government identified 5 ‘transformatory technologies’ that would ‘personalise care, improve outcomes, increase productivity and boost economic growth’ in NHS England (Department of Health and Social Care). The first technology on that list was ‘data’. In the Scottish Government’s (2021) Digital Health and Care Strategy it said ‘If digital is the engine of our health and care system, data is its fuel.’. Regardless of how we take this —in our opinion, that is a big ‘IF’— data is clearly regarded as a vital thing for contemporary healthcare.

In this blog, we ask a simple question: where are all these data (or, if you are less old fashioned — where is all of this data)? It feels a little absurd to write this in our current moment when data appears to be everywhere, but there is an important kernel within this slightly strange question. Data lives in machines, on paper, and —perhaps— in the mind. But first it comes from somewhere. If you want to know about the world, you have to get data. You do some work to “get” data “out” of the world. The past decades of great data abundance have rather obscured the spatial aspects of this problem. Data live inside computers. It is statistical fodder to be thrown at problems; data is everywhere, one can simply reach out and pluck the bits you need. Abundance obscures the peculiarity of a thing, and the peculiarity of data is its ability to stretch and pass between place and between context. It comes from somewhere, and it gets taken elsewhere, that’s its fundamental condition of being. To put it rather crudely: It gets taken, it gets placed, and then it gets used.

Where is data?

Ethnomethodologically inclined social scientists are a useful way of thinking this through. For an ethnographer, collecting data usually means going somewhere. Of course, this is as true for the bench scientist who must go to their laboratory as it is for the ethnographer. However, the laboratory creates a great abundance of data, and thus, its strangeness is supressed in this productivity (see: Woolgar & Latour, 1986). Ethnographers go to places (plural); different places. If they want to know about a culture, they try to get into it. There is all manner of interesting and well litigated discussion about how one might go about getting into a culture, whether it is possible to get in as an outsider, and whether it is more useful to be inside or outside in this regard. We can glance at those briefly, but here, for now, let us simply concentrate on the importance of place in all of this.

Regardless of whether they consider it more useful to observe or to partake, ethnographers must be in the place they want to study. Because of this, ethnographers tend to think a lot about space, place, and the nature of ‘sensing’ when collecting data (See: Massey 2005; Pink 2015; Ingold 2016). The place, its inhabitants, the milieu of data at the point of its collection are each vitally important to the ethnographer when thinking through data. That so much ink has been spilled by anthropologists and social scientists about the senses, about culture, about what it means to ‘participate’ and to ‘extract’ data from the world, should not distract us from the fact that such considerations hold true for the other sciences too. Indeed, it holds for anyone engaged in knowing about the world. For example, for a doctor taking a blood pressure reading, it matters where the blood pressure is taken. If a blood pressure reading is taken as part of a GP appointment, outside of the patient’s normal environment, with all of the stresses and sensory exchanges that come with attending a medical appointment, the reading is liable to be deceptively high. The data point “SBP/DBP” is generally understood as an objective measure, however, we must understand it as a product of its place. White coat hypertension is now such a widely recognised phenomena that it now has its own International Classification of Disease (ICD) code (MC800.00). All data comes from a place, Marc Berg and Els Goorman (2002) suggest the ‘law of medical information’, which suggests that understanding the context of data production is vital for its later use, and that the repurposing of data is itself a labour that should be understood. That is, even if we standardise datasets, we are producing data anew, and we are producing it anew in a place that we would do well to understand. 

What interests us in these questions about data, is both the epistemological and the practical problem of place. Clearly there are effects of place on data produced but also, when collecting data, ethnographers must find their “informants”, they must negotiate access, speak to “gatekeepers” who might introduce them to the field, authorise them to be there, introduce them to their informants. They might have to live in a different place for a long time, eat differently, exercise differently, live differently as a result of their data collection. Do these problems remain when we’re working with pre-built data sets?

This all brings us to Trusted Research Environments (TREs). TREs promise researchers access to data from another place, but we very rarely consider TREs as places themselves. Places that must be accessed, places that have their own culture, their own milieu, their own topography. What happens when we consider where TREs are?

Where are TREs?

TRE’s allow researchers to access data from multiple environments in a single ‘place’ that is secure and operated by data controllers. Accessible by sufficiently trained and approved researchers, the data held in a TRE can only be analysed within the secure environment, and is not able to be extracted to an external location. For digital researchers such as data scientists wanting to access data, the TRE acts as the place in which their ‘informants’ are located, whom can only be accessed through the TRE gatekeepers. What, then, is the relationship between the TRE’s place, and the ‘field’ for which it holds data about? 

Like ethnographic fieldwork, there are still requirements needed to gain access to the ‘field’ that researchers must abide by to be able to conduct their research and analysis. Though TRE’s are digital environments, there are regulations about the physical places that researchers can be in when accessing their ‘informants’ (the data). Researchers are not allowed to use TRE’s in public spaces to protect confidential data from those not permitted to see it. However, TRE’s do allow researchers to access their ‘field’ of research without having to enter the physical space, making access to data and ‘informants’ easier than arranging physical visitation to places such as hospitals or community healthcare facilities. More researchers are able to access data, allowing for more efficient and collaborative research that has the possibility to contribute to impactful studies and improve patient outcomes. Furthermore, as disciplines such as data science in healthcare become increasingly popular, the need for these researchers to visit the spaces from which the data they need is collected is growing smaller. This is a fundamental shift in the relationship between researchers, their data, and place, as accessing the ‘field’ is no longer required to access the ‘data’. 

Of course, not all research needs to have physical locations, or physical ‘informants’, and in data science this is not a reasonable expectation. TRE’s allow valuable research and analysis to be undertaken in safe ‘places’ that otherwise would not be possible. Nonetheless, the changing relationship between data and place poses interesting questions about local knowledge and context found within the field, and the impact this may have on the data. What kind of implications does this have for healthcare research? Are digital researchers becoming too removed from the ‘source’ of their data (the patient), and does this distance present itself in findings or results? Does it matter that the gatekeepers and mediators of a TRE may not have this local knowledge, and are relying on the researcher to understand their data on their own? These questions require further enquiry, and do not mean to suggest that work undertaken within a TRE is any less valuable than that conducted within a physical environment. 

*

So where are data? We opened by saying this was a slightly absurd question, and indeed, we maintain that it is. To return to the slightly unfashionable tensing of the word “data”, we think that it is more useful to ask where is this datum? That is because data travels, it is relocated, reshaped, and re-cultured every time it passes from place to place. Do all of us need be ethnographers to speak about the world, to go and put ourselves in the place that the data comes from? This, it seems to us, is to slightly miss the point. TREs hold such potential because they allow researchers the ability to work at scale, to make use of data collected elsewhere. With that said, when working with data from a TRE, it is useful to understand how the TRE works. To understand how it gets its data, how it makes it ‘research ready’, and how and through what tools TREs make that data available to researchers. You might not need to go to the hospital and watch the blood pressure reading of each patient in clinic to determine whether the ICD code used was correct or not, however, you might hope that your colleagues in the TRE had enough domain knowledge (or, we might say ‘ethnographic sense’) to determine whether doctors in that clinic knew to use MC800.00 for white coat hypertension.

REFERENCES

Berg, M. And Goorman, E. (1999) The Contextual Nature of Medical Information, International Journal of Medical Informatics, 56(1), pp. 51–60. doi: 10.1016/S1386-5056(99)00041-6.

Department of Health and Social Care (2025) 10 Year Health Plan working groups. Department of Health & Social Care. [Online] Available online at: https://www.gov.uk/government/groups/10-year-health-plan-working-groups (Accessed: 26 May 2026).

Digital Health and Care Directorate (2021) Digital health and care strategy. Digital Health and Care Directorate. [Online] Available online at: https://www.gov.scot/publications/scotlands-digital-health-care-strategy/pages/9/ (Accessed: 26 May 2026).

Ingold, T. (2016) Lines : a brief history. London ; Routledge.

Latour, B. and Woolgar, S. (1986) Laboratory life : the construction of scientific facts. [Second edition] / with a new postscript and index by the authors. Princeton, N.J. ; Princeton University Press.

Massey, D. (2005) For space. London: Sage.

Pink, S. (2009) Doing Sensory Ethnography. London: Sage