Reviewing qualitative GIS research—Toward a wider usage of open‐source GIS and reproducible research practices

Abstract

AbstractGeographic Information Systems (GIS) have become an indispensable tool in all fields dealing with geographic data in academia as well as the public and private sector. After some initial reservations, human geography and other social sciences have also embraced GIS technologies especially by extending and adjusting the available data models for qualitative data. The result was coined qualitative GIS. Given the diversity and fragmentation of the research in which qualitative GIS have been applied, a comprehensive literature review is overdue. Therefore, we retrieved all articles that dealt with qualitative GIS from the ISI Web of Science. From the resulting 380 relevant manuscripts, we extracted study‐specific information and analyzed it by means of descriptive statistics, text mining, detrended correspondence analysis (DCA), and k‐means clustering. Our results reveal four major thematic domains within which qualitative GIS research takes place, namely, “conceptual contributions to qualitative GIS,” “theory and empiricism of public participation GIS,” “health, youth, and the urban environment,” and “ecosystem services, landscapes, and tourism”. Some of these trends are popular in specific regions (e.g., “ecosystem services, landscapes, and tourism” research in Australia). Furthermore, we found that a large body of studies lacks clear information on the used software packages and that the share of free open‐source GIS software is negligible. Overall, we would encourage more transparency, a wider use of open‐source software, and interdisciplinary collaboration to even further advance and promote qualitative GIS research. To put our recommendations into practice, we made the data used in this manuscript and the corresponding analysis publicly accessible (https://github.com/jannes‐m/qual_gis). To further facilitate the replication of our results, we exclusively used open‐source software, namely, R, leaflet, and PostgreSQL. These and other open‐source software packages, for example, Python and QGIS, might have the potential to fundamentally change the way to conduct qualitative GIS research. This is equally true for methods borrowed from other fields such as ordinations, cluster algorithms, and interactive web mapping, which might be promising tools in human geography and the social sciences.