From past to present: understanding the impact of sampling bias on data recorded by the portable antiquities scheme
From past to present: understanding the impact of sampling bias on data recorded by the portable antiquities scheme
This thesis explores the distribution of the data collected by the Portable Antiquities Scheme (PAS), which records archaeological artefacts recovered by members of the public, many of whom are metal detector users. With over 780,000 artefacts recorded onto an online database, it is not surprising that the PAS data have become an essential part of many research projects. However, until now there has been little work on the biases that are inherent in the data, in particular the effect of sampling bias on the distribution of finds. The thesis is grounded in discussions of sampling theory and collection bias, and suggests that bias can enter the archaeological record in seven stages (burial/loss, preservation, survival, exposure, recovery, reporting, and recording). A range of the factors contained within these stages are explored through three study areas (the Isle of Wight, Hampshire, and Northamptonshire). Within each area the distribution of PAS artefacts is first compared to the known distribution of past human activity, and then to a range of physical and man-made landscape features. Six case studies are used to explore specific elements in more detail, to focus the analysis on those factors that have the most impact on the distribution of finds. These statistical and spatial analyses are supported by a survey of metal detecting methods, which uses primary and secondary sources to explore the impact of different attitudes, experience, and techniques on the distribution of finds. This research identifies a series of choices made by collectors, recorders and landowners, that are shown to have most influence over the distribution of PAS finds – these range from ‘where to search’, through ‘what to recover’, to ‘what to report’. This research shows how an understanding of these choices is essential for those wanting to incorporate the PAS data into their research.
Robbins, Katherine
c0ee02ed-39bc-4ec5-829c-6a86f933b1c9
2012
Robbins, Katherine
c0ee02ed-39bc-4ec5-829c-6a86f933b1c9
Bland, Roger
762bcf55-1844-4add-8689-2ba8b1ee1b34
Earl, Graeme
724c73ef-c3dd-4e4f-a7f5-0557e81f8326
Robbins, Katherine
(2012)
From past to present: understanding the impact of sampling bias on data recorded by the portable antiquities scheme.
University of Southampton, Archaeology, Doctoral Thesis, 733pp.
Record type:
Thesis
(Doctoral)
Abstract
This thesis explores the distribution of the data collected by the Portable Antiquities Scheme (PAS), which records archaeological artefacts recovered by members of the public, many of whom are metal detector users. With over 780,000 artefacts recorded onto an online database, it is not surprising that the PAS data have become an essential part of many research projects. However, until now there has been little work on the biases that are inherent in the data, in particular the effect of sampling bias on the distribution of finds. The thesis is grounded in discussions of sampling theory and collection bias, and suggests that bias can enter the archaeological record in seven stages (burial/loss, preservation, survival, exposure, recovery, reporting, and recording). A range of the factors contained within these stages are explored through three study areas (the Isle of Wight, Hampshire, and Northamptonshire). Within each area the distribution of PAS artefacts is first compared to the known distribution of past human activity, and then to a range of physical and man-made landscape features. Six case studies are used to explore specific elements in more detail, to focus the analysis on those factors that have the most impact on the distribution of finds. These statistical and spatial analyses are supported by a survey of metal detecting methods, which uses primary and secondary sources to explore the impact of different attitudes, experience, and techniques on the distribution of finds. This research identifies a series of choices made by collectors, recorders and landowners, that are shown to have most influence over the distribution of PAS finds – these range from ‘where to search’, through ‘what to recover’, to ‘what to report’. This research shows how an understanding of these choices is essential for those wanting to incorporate the PAS data into their research.
Text
__soton.ac.uk_ude_personalfiles_users_slb1_mydesktop_Robbins thesis.pdf
- Other
More information
Published date: 2012
Organisations:
University of Southampton, Archaeology
Identifiers
Local EPrints ID: 360475
URI: http://eprints.soton.ac.uk/id/eprint/360475
PURE UUID: 847aa61f-78f9-4c35-8ad1-a94a6a301364
Catalogue record
Date deposited: 10 Dec 2013 14:53
Last modified: 14 Mar 2024 15:38
Export record
Contributors
Author:
Katherine Robbins
Thesis advisor:
Roger Bland
Thesis advisor:
Graeme Earl
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics