There has been a growing use of geographic information due to recent technological developments. Many people outside GIS and other spatial disciplines are increasingly using geographic information for day-to- day decision making (Kraak and Molenaar, 1997:815). This trend has raised concern for delivery of high quality and accurate geographic information by data producers. Geographic data are usually supplemented with metadata to help users discover and ascertain suitability of the datasets for a particular task. However, studies have shown that metadata delivered by data producers have not been useful to guide users in discovering fitness for use of such data (Onchaga, 2003: 6). Metadata in most cases are presented in complex ways that tend to confuse users particularly non experts users and as such most users do not pay much attention to metadata.
Current metadata is based on various standards, using common specifications to describe different data types (for example, the ISO 19115 standards outlined in ISO /TC211, 2009:54). It should also be user focused, with formal procedures and tools to guide end users, detailing information on the uncertainties and risks associated with using the datasets (Agumya and Hunter 1996: ). Metadata by data producers can be communicated as decision support information rather than as way of protection against any potential litigations as presently being perceived by end users (Devillers and Jeansoulin 2006: 240). Mind map illustrating main issues in the research is given in Appendix 1.
This study aims to identify how users interact with metadata and how geographic data producers can better present metadata so as to make them more meaningful to end users. Specifically, the study will;
- Examine how users of spatial datasets currently utilise metadata provided by data producers.
- Identify ways metadata can be communicated to the users of geographic information. For example, aside from textual metadata, are there other ways data producers can communicate metadata to different types of customers.
- Compare Ordnance Survey metadata materials with metadata standards such as ISO 19115, FGDC.
This study will be carried out using a combination of two approaches. Firstly, current literature will be reviewed and used to identify different ways metadata can be presented and communicated to the end users. The second aspect involves the use of questionnaires survey to elicit responses from users of geographic information in academia, local authorities, land registry, NHS in London. 60 structured questionnaires will be administered via mailing lists. Pilot test of the questionnaires will be conducted by administering the questionnaires to five (5) UCL M.Sc GIS students to ascertain the reliability of the data and ease of analysing the data. The questionnaires will ascertain from the users how the current metadata helps them in assessing fitness for purpose and what kind of decisions they make when confronted with uncertain data. Information on the completeness of datasets, currency of the data, semantic accuracy and other metadata issues will also be sought through the questionnaires. The design of the questionnaire will be made to cover all the objectives of the study and address all aspects of the research problems.
The responses elicited from the questionnaires will be subjected to both qualitative and quantitative analyses manually. Simple descriptive statistics such as percentages, tables, proportions and maps will be used to analyse the results of data where appropriate.
Metadata simply means data about data. Quality information regarding geographic datasets is presented in form of metadata. "Metadata provides information on sources, data quality, data extent, content, production processes and responsibilities to users of geographic data" (Shi et al, 2002:211). These information on geographic datasets by data producers allows users to assess the fitness for purpose of geographic datasets. Geographic datasets can be evaluated to be fit for use when the consequence of error and uncertainty associated with such datasets is deemed acceptable to the user for that particular application (Agumya and Hunter, 1999:35). Agumya and Hunter (1999:35) further posited that fitness for use of geographical data can be assessed using standard-based procedure and risk - based procedure. Current metadata is based on various standards published by different organisations such as ISO 19115, Content Standard for Digital Geospatial Metadata Data by Federal Geographic Data Committee (FGDC). These standards ensure consistency in contents and style within a particular application area and community (SDI Cookbook, 2004: 28). Users are able to make useful comparison among datasets using standardized metadata materials.
Vasseur et al (2003:7) canvassed using ontological approach for evaluating fitness of purpose of geospatial data by comparing user's requirements (problem ontology) and data specifications (product ontology) using several quality criteria.
However, geographic data users particularly novice users, find it difficult evaluating usability of datasets for specific tasks because of the way metadata is being communicated. Most data producers end up communicating metadata as a protection from potential litigations rather than as a decision support tool (Devillers and Jeansoulin, 2006:241). Devillers and Jeansoulin, (2006:240), further posited that textual metadata may not be the most efficient way to communicate quality information on geographic datasets. According to Shi et al, (2002:141), Devillers and Jeansoulin, (2006:241), visualization has been one of the effective approaches to communicate information about geographic data quality and uncertainty. Using graphical methods, complex and voluminous datasets can be communicated to users thus enhancing visual comprehension and aesthetic appeal. Further research has also addressed the issue of user interaction issues in visualizing data quality, thus visualizing metadata (Devillers and Jeansoulin, 2006:242). According to Devillers and Jeansoulin, (2006:242), graphic and cartographic design should be considered in light of user cognitive and perception abilities.
There may be possibility of having low response rate of questionnaires from the targeted population. This problem is particularly inherent in field based researches. Some techniques will be adopted so as to maximise the response rate. For example, reminder mails with a copy of the questionnaires will be sent out intermittently to respondents who have not returned the questionnaires. Other follow-up mechanisms, like telephone calls (for those respondents the researcher have access to their telephones - eg fellow students) will be adopted so as increase the response rate.
Expected project outcomes:
- The research will determine whether metadata by data producers is presented in such a way as to guide users of the datasets in assessing fitness for purpose.
- The project will examine whether users of geographic data consider metadata before using such datasets.
- The research will then come up with standard and best practice way to communicate metadata including information on the associated risks of using datasets.
As this is a sponsored research project, planned deliverables to the industrial collaborator, Ordnance Survey include:
- Project report - including recommendations as to how to improve metadata delivery
- Executive Summary - summary report of the project.
Project planner (Gantt chart) illustrating the tasks and resources to execute the above research plan is presented in Appendices 2 and 3.
The link below contains list of Ordnance Survey sponsored M.Sc research projects for 2009/2010
- "Metadata for Geographic Information" is number six (6) in the list.
- Agumya, A., and Hunter, G., (1999); A Risk-Based Approach to Assessing the 'Fitness for Use' of Spatial Data: Urisa Journal 11(1) Spring 1999: Pp.35, Available from http://www.urisa.org/files/Vol11No1.PDF: [Accessed online 08/02/2006]
- Agumya, A., and Hunter, G., (1996); Assessing Fitness for Use of Spatial Information Usage and Uncertainty Decision - Proceedings of the GIS/LIS Annual Conference and Exposition, 1996. Available from http://training.esri.com/campus/library/Bibliography: [Accessed online 08/02/2010].
- Devillers, R., and Jeansoulin, R., (eds.) 2006; Fundamentals of Spatial Data Quality: GIS Series, Pp.237-245 ISTE Limited, 2006.
- IDRC (2003); Designing and Conducting Health systems Research Projects: Proposal Development and Fieldwork; Pp. 3 - 27: Available from http://www.idrc.ca/en/ev-56581-201-1-DO_TOPIC.html [Accessed online 07/04/2010]
- Kraak, M., and Molenaar, M., (eds.) 1997; Advances in GIS Research 11: Proceedings of the Seventh International Symposium on Spatial Data Handling (SDH 96): Pp.815 - 816,Taylor and Francis, 1997.
- ISO /TC211 (2009); Standards Guide - ISO/TC211 Geographic Information and Geomatics, Pp. 54-56, Available from http://www.isotc211.org/Outreach/ISO_TC%20_211_Standards_Guide.pdf [Accessed online 18/02/2010].
- Onchaja, R., (2003); Extending the Quality Concept in Geo-Information Processing, Journal of GIS
- Development, 2003, Pp.1-10 Available from http://www.gisdevelopment.net/applications. [Accessed online 08/02/2010]
- SDI Cookbook (2004); Developing Spatial Data Infrastructure - SDI Cookbook, Version 2-0, 2004, Pp. 24 - 94, Available from http://www.gsdi.org/docs2004/Cookbook/cookbookV2.0.pdf [Accessed online 08/04/2010]
- Shi, P., Fisher, P., Goodchild, M.,(eds.) 2002: Spatial Data Quality: Taylor and Francis 2002, Pp.140 - 214
- Vasseur, B., Devillers, R., Jeansoulin, R.,(2003); Ontological Approach of the Fitness of Use of Geospatial Datasets: Proceedings of the 6th AGILE, April 24 - 26 2003, Lyon, France: Pp.7, Available from http://plone.itc.nl/agile_old/Conference/lyon2003/proceedings/58.pdf [Accessed online 07/04/2010]