Developments in computational analysis capabilities and growing access to georeferenced digital data broaden the horizon of geographical research and practice. In this context, interest in spatial analysis across the sciences and humanities [26, 30] as well as in policy making  has been growing in the last two decades. In particular, the spread of georeferenced data practices can enable diverse interventions and decision making based on maps and the analysis of places (defined here as situated practices, activities and events). For example, geographic analysis can inform local, regional, national and international development, community justice and community psychiatry practices. Understanding ourselves in spatio-temporal context also underpins our dialogues with diverse communities.
This project aims to enhance our ability to reflect on the consequences of using geographic modelling and analysis tools in diverse interpersonal, institutional, political and cultural contexts. Understanding the values and discourses associated with such data practices is more and more important, given the growing popularity of geographic analysis in diverse contexts. The project aims to contribute to existing approaches that study the assumptions and values embedded in and the consequences of such data practices through empirical case studies.
Geography has a long tradition of studying and representing places (situated activities and events). Over the course of its history, the discipline developed research traditions that rest on diverse ontological and epistemological assumptions. For example, approaches originating in the ‘quantitative turn’ of the 1950s and 1960s developed tools and concepts to study the spatio-temporal unfolding of social processes using concepts and methods borrowed from physics and mathematics [cf. 1]. In addition, critical geography and cartography developed a range of concepts and methods that help reflect on the experiential, political, aesthetic, affective and metaphysical aspects of places as well as geographic and cartographic (research) practices. Such diverse approaches can bring forth conceptions of places as systems that can be planned, managed and controlled [cf. 15]. They can also help understand the politics, affects and values associated with places  that are related to quality of life  and the situated experiences of people and communities . Current geographical meaning-making (i.e. research and practice) can provide opportunities for these diverse traditions of thought to meet.
Geographic data analysis can form part of scientific , governmental , commercial  and non-governmental  practices. Moreover, thanks to the increasing popularity of geographic analysis and capabilities afforded by digital tools, a growing number of actors (e.g. groups of people, institutions) can be involved in geographical analysis, both as researched and as researching subjects. For example, places can be studied through the analysis of georeferenced social media data that capture interpersonal interaction [6, 12], georeferenced news reports that capture situated activities and framings  and volunteered geographic information that can capture local knowledges . Moreover, computational techniques can be used by increasingly diverse audiences as participatory agent based modelling  and crime data hackathons  illustrate. The diversity of these domains and the actors involved in geographic meaning-making illustrate the diversity of geographic and social imaginations (i.e. the ways in which we imagine, plan, understand, manage, and enact our social and material environment) that georeferenced data practices enable.
Sociological studies of science [19, 29] and finance [e.g. 20] have illustrated that data analysis, modelling and their interpretation are embedded in and contingent on their interpersonal, institutional contexts. For example, the avalanche of printed numbers enumerating people for military and taxation purposes (among others), statistical reasoning and the notion of probability co-evolved with Western conceptions of community, society and individual . Similar to this, the availability of audio-recording and a research tradition focusing on the orders and meanings of face-to-face interaction in the 1960s enabled the development of Conversation Analysis, which posed novel ways of investigating, interpreting and questioning intersubjective engagement and societal order .
Inspired by such accounts it can be argued that while data practices enable geographic and sociological imaginations, the relationship between data practices and the discourses they enable cannot be studied from a purely techno-scientific point of view. Rather, such practices need to be understood as assemblages (context-dependent and dynamic collections) of human and non-human actors  (for example, computers, data analysis tools, researchers, research laboratories, universities, publications, conferences etc.) [cf. 23]. Understanding the unfolding of such assemblages calls for an approach that can account for diverse (e.g. political, scientific, symbolic, aesthetic, embodied) aspects of geographic meaning making. In line with this, the current project aims to provide a thick description of data practices that aim to understand places, situated activities and events.
The state of the art research questions were identified based on a review of relevant literature. However, the characteristics of the case studies (TBC) will further inform the focus of the project.
Research Gap 1: A small percentage of the papers that reflect on the values associated with and consequences of data practices study (geographic) data analysis processes empirically. Among the empirical studies, most papers reflect on computational data analytics from the perspective of the researcher who uses such techniques, rather than investigating data practices from an ethnographic perspective. [cf. 2] An ethnographically informed study of geographic research can help reflect on the link between data analysis practices and conceptions of place in detail. As ,  and  illustrate, the relationship between data and places is multi-faceted. For example, data can refer to places, constituted by situated practices, and be analysed, visualized and categorised in multiple ways that afford different ways of narrating spatio-temporal relationships. Moreover,  compares ‘close’ and ‘distant’ readings of data associated with the Arnold Arboretum of Harvard University. By focusing on the meaning of variables, he illustrates that a cursory analysis of data limits the types of stories that can be told about the Arboretum (e.g. important historical events in the life of the Arboretum can only be inferred through attending to local stories in addition to data analysis). Moreover, by creating a range of data visualizations, he narrates the history of the Arboretum in ways that would otherwise be difficult to imagine and depict.
Research Question 1: How are places and socio-spatial processes understood through digital data analysis practices?
Research Gap 2: As mentioned above, geography developed diverse methods to study places. However, the disciplinary war between ‘quantitative’ and ‘qualitative’ (these terms refer to historically significant and classification which is still in use, but the validity and meaning of which has been topic of lively debates) geography (that dates back to the 1960s and 1970s) still haunt the discipline. In 2013, the Economic and Social Research Council called for creating dialogues between quantitative and qualitative methods in geography  cited by .
Research Question 2: How do existing research practices establish relationships between ‘qualitative’ and ‘quantitative’ thinking?
Research Gap 3: Cities can be understood as dynamic assemblages of spatio-temporally unfolding relational domains, such as neighbourhoods, institutions, or more or less cohesive residential groupings. Borders and bordering areas are in a constant state of becoming, such as through high level governmental decision making (‘macro-level’) (e.g. defining the boundaries of neighbourhoods and census tracts) as well as the day to day interaction between people (‘micro’ level)  and at the same time, borders also constrain the practices that constitute them. Moreover, borders can be porous, rigid, fluid .  argues that the study of borders on the scale of cities and neighbourhoods have been neglected in favour of investigating “urban splinters and enclaves, ethnic communities and social classes, or other more or less clearly defined and seemingly self-contained urban entities’’ (p. 4.). In particular,  argues that understanding the urban condition requires the joint study of macro and micro-level constraints that constitute borders.
Research Question 3: How do the case studies assume, create and narrate borders through diverse research methods?
The project will study a few relevant research practices through an ethnographically informed qualitative case-study method. The case studies will be confirmed on an on-going basis. The method will be informed by previous research that developed concepts and tools to study the journey of data between different locations , the way modelling is embedded in institutional and political contexts , the construction of scientific arguments  and the culture of laboratory practices .
As outlined in the first paragraph, the project can enhance our capability to assess the implications of geographic data practices in diverse political, cultural and institutional contexts. Moreover, it can also help develop further perspectives on interdisciplinary research [e.g. 3]. Finally, it can contribute to our understanding of emerging cartographic practices (e.g. mapping social media, participatory mapping) and digitally mediated situated experiences.
This author is supported by the Horizon Centre for Doctoral Training at the University of Nottingham (RCUK Grant No. EP/L015463/1) and Ordnance Survey.