The Impact of Human Decision-making on the Research Value of Archaeological Data

‘Good data in, good data out’ is a well-known adage, often used to consider the ‘precision’ of data entry rather than other human activity when inputting information into a database. These could be omissions or errors, but are just as likely to be inconsistencies in data recording. This is especially relevant when thinking about the efficiencies that recorders might make (i.e., to make their work easier) and what this could mean for end users (e.g., researchers). For example, a recorder of an archaeological object might do things differently based on their knowledge or understanding of the value of recording certain types of data (e.g., the various characteristics of the object) and how this information is likely to be used by others. With this in mind, this paper will consider the research question ‘how can advancing FAIR principles be used to contextualize the impact of human decision-making on the research value of archaeological data?’, taking (as a test case) the Portable Antiquities Scheme (PAS) database of archaeological finds from England and Wales. The methodology used here examines data recorded within the PAS database directly and through a separate web application demonstrator (PASampo) that is being developed. The latter, in particular, enables ease-to-use visualization of the data to provide an analytical evaluation of how data is inputted into the PAS database and, more significantly, how the researcher might extract this data. Through a series of case studies exploring aspects of material culture, we will highlight how data quality is affected by human decision-making alongside elements of the database framework. We will then argue that this provides the basis for new learning alongside adopting FAIR principles that could benefit the development of a wide range of future archaeological and cultural heritage databases, including the PAS database itself.