Sociotechnical Obstacles to Archaeological Data Reuse

Sobotková, Adéla

doi:10.1017/aap.2017.37

“…Calls for archaeologists to improve data quality for large-scale research have been made for at least the past two decades (Doerr et al 2004;Kansa and Kansa 2011;Faniel et al 2013;Austin 2014), but limited progress has been made (Kintigh et al 2014;Sobotkova 2018). Concerns over the transparency and reproducibility of archaeological research are more recent (Marwick 2017b), but archaeology is unlikely to avoid the reproducibility crisis so profoundly affecting other disciplines.…”

Section: Discussionmentioning

confidence: 99%

Introducing Preregistration of Research Design to Archaeology

ROSS¹,

BALLSUN-STANTON²

2022

Digital Heritage and Archaeology in Practice

0

View full text Add to dashboard Cite

Archaeology has an issue with "just-in-time" research, where insufficient attention is paid to articulating a research design before fieldwork begins. Data collection, management, and analysis approaches are under-planned and, often, evolve during fieldwork. While reducing the amount of preparation time for busy researchers, these tendencies reduce the reliability of research by exacerbating the effects of cognitive biases and perverse professional incentives. They cost time later through the accrual of technical debt. Worse, these practices hinder research transparency and scalability by undermining the quality, consistency, and compatibility of data. Archaeologists would benefit from embracing the "preregistration revolution" sweeping other disciplines. By publicly committing to research design and methodology ahead of time, researchers can produce more robust research, generate useful and reusable datasets, and reduce the time spent correcting problems with data. Preregistration can accommodate the diversity of archaeological research, including quantitative and qualitative approaches, hypothesis-testing and hypothesis-generating research paradigms, and place-specific and generalizing aims. It is appropriate regardless of the technical approach to data collection and analysis. More broadly, it encourages a more considered, thoughtful approach to research design. Preregistration templates for the social sciences can be adopted for use by archaeologists.

show abstract

“…Calls for archaeologists to improve data quality for large-scale research have been made for at least the past two decades (Doerr et al 2004;Kansa and Kansa 2011;Faniel et al 2013;Austin 2014), but limited progress has been made (Kintigh et al 2014;Sobotkova 2018). Concerns over the transparency and reproducibility of archaeological research are more recent (Marwick 2017b), but archaeology is unlikely to avoid the reproducibility crisis so profoundly affecting other disciplines.…”

Section: Discussionmentioning

confidence: 99%

“…In the future, stakeholders will expect archaeologists' data and code, not just our results and conclusions. At present, however, data repositories are underpopulated and datasets are not routinely reused (McNutt et al 2016;Sobotkova 2018). This underpopulation inhibits reproduction or verification of results, independent analyses of primary data, and the application of new techniques to old datasets.…”

Section: The Problem Of "Just-in-time" Archaeologymentioning

confidence: 99%

Introducing Preregistration of Research Design to Archaeology

Ross¹,

Ballsun-Stanton²

2021

Preprint

View full text Add to dashboard Cite

Archaeology has an issue with "just-in-time" research, where insufficient attention is paid to articulating a research design before fieldwork begins. Data collection, management, and analysis approaches are under-planned and, often, evolve during fieldwork. While reducing the amount of preparation time for busy researchers, these tendencies reduce the reliability of research by exacerbating the effects of cognitive biases and perverse professional incentives. They cost time later through the accrual of technical debt. Worse, these practices hinder research transparency and scalability by undermining the quality, consistency, and compatibility of data. Archaeologists would benefit from embracing the "preregistration revolution" sweeping other disciplines. By publicly committing to research design and methodology ahead of time, researchers can produce more robust research, generate useful and reusable datasets, and reduce the time spent correcting problems with data. Preregistration can accommodate the diversity of archaeological research, including quantitative and qualitative approaches, hypothesis-testing and hypothesis-generating research paradigms, and place-specific and generalizing aims. It is appropriate regardless of the technical approach to data collection and analysis. More broadly, it encourages a more considered, thoughtful approach to research design. Preregistration templates for the social sciences can be adopted for use by archaeologists.

show abstract

“…First, preserving data for reuse and actually reusing other people's data isn't given the priority it deserves in our discipline (Sobotkova 2018). In the EAFWG project, it was not possible to locate and incorporate all the relevant faunal data from the interior East within our three-year schedule.…”

Section: Creating Accessible Database Collectionsmentioning

confidence: 99%

Zooarchaeological Database Preservation, Multiscalar Data Integration, and the Collaboration of the Eastern Archaic Faunal Working Group

Neusius

¹

,

Styles²,

Peres³

et al. 2019

Adv. archaeol. pract.

View full text Add to dashboard Cite

Data preservation, reuse, and synthesis are important goals in contemporary archaeological research that have been addressed by the recent collaboration of the Eastern Archaic Faunal Working Group (EAFWG). We used the Digital Archaeological Record (tDAR) to preserve 60 significant legacy faunal databases from 23 Archaic period archaeological sites located in several contiguous subregions of the interior North American Eastern Woodlands. In order to resolve the problem of synthesizing non-standardized databases, we used the ontology and integration tools available in tDAR to explore comparability and combine datasets so that our research questions about aquatic resource use during the Archaic could be addressed at multiple scales. The challenges of making digital databases accessible for reuse, including the addition of metadata, and of linking disparate data in queryable datasets are significant but worth the effort. Our experience provides one example of how collaborative research may productively resolve problems in making legacy data accessible and usable for synthetic archaeological research.

show abstract

Sociotechnical Obstacles to Archaeological Data Reuse

Cited by 32 publications

References 15 publications

Introducing Preregistration of Research Design to Archaeology

Introducing Preregistration of Research Design to Archaeology

Introducing Preregistration of Research Design to Archaeology

Zooarchaeological Database Preservation, Multiscalar Data Integration, and the Collaboration of the Eastern Archaic Faunal Working Group

Contact Info

Product

Resources

About