Development of Metadata Education Strategies

for University Curriculum in GIS/GIScience

Final Report, January 2000

Margo Berendsen

Spatial Data and Visualization Center

University of Wyoming

Contents:

  1. Background
  2. Importance of integrating metadata into GIS curriculum
  3. Issues involved in curriculum development and pedagogy
  4. Review of GIS-related courses: content and materials
  5. Synthesis of recommendations and conclusions


1. Background

Many advantages have been identified and realized in the implementation of metadata development in various sectors of the geospatial community, especially metadata in a standard form, such as the content standard endorsed the Federal Geographic Data Committee (FGDC). However, because of the complex and variable nature of geospatial data, the standard developed to document this data is necessarily complex, as well as time-consuming to learn. The purpose of this project, sponsored by the FGDC and the University Consortium for Geographic Information Science (UCGIS) is to:

In order to accomplish these goals, a meeting was organized to bring GIS education and metadata experts together to discuss educational and content issues specific to the development of educational material for metadata. The meeting was held on June 3-5 1999 in Herndon, Virginia with 25 participants from universities and organizations around the country involved in GIS and metadata education. The first day of the meeting focused on presentations of different metadata training efforts and GIS education methods to provide a background of the target audiences, educational settings, and educational content. On the following day, participants formed three working groups to discuss educational objectives and existing/potential constraints for these settings: traditional university curriculum, self-paced or distance-learning environment and professional workshop/short course environment.

The first aspect in developing educational objectives for all three working groups was to identify different audiences. The spectrum of GIS audiences ranges from the "informed user" who wants or needs only a basic awareness of spatial concepts; to the "GIS user" using GIS as a tool within another discipline; the "GIS analyst" with a more focused interest in GIS as an information tool; and finally the "GIS developer". The prevalence of spatial data products and easy-to-use desktop mapping and GIS software in today’s society is also creating a new audience, "the novice" who may not have any awareness of basic spatial concepts at all but still has access to spatial tools. The "novice" is a particularly difficult audience to reach, since they may not seek any formal educational experience at all or if they do it is likely to limited be either in time or in method. Therefore the importance of alternative learning mechanisms – self-paced, easily accessible tutorials or distance learning programs/workshops is not to be under-rated.

Educational objectives, arranged by target audience for the university curriculum and self-paced/distance-learning curriculum, turned out to be very similar, though the instructional delivery methods are somewhat different. There was a general consensus that the main objective in these two environments is to weave metadata education into existing curriculum and educational materials in all GIS-related courses as well as related fields such as remote sensing and GPS. It should not be treated as a separate component, but as an integrated part of the entire GIS process, therefore it should also be an integrated component of a GIS class instead of just one lecture or part of a lecture.

Educational objectives for workshops were different because of the existence of workshops devoted solely to metadata training, a much more focused subject than the varied content of most GIS-related courses. In this setting however, educators may encounter "GIS professionals" without sufficient foundational knowledge in spatial concepts such as scale, projections, and representation. Because of the relative "newness" of GIS technology and scarcity of professionals fluent both in their native field and in spatial techniques, this audience is probably quite prevalent. In such cases, metadata training workshops can play a crucial role in helping to disseminate fundamental spatial concepts to the workforce, since many elements of the metadata content standard (data quality, spatial organization, spatial reference) are directly related to these fundamental concepts. Use of metadata in this respect promotes a more responsible and informed approach to creating, using and maintaining data in addition to the other well-recognized benefits of metadata such as protecting data investments and facilitating data sharing. The basic problem still remains, however, as to how to effectively teach an unwieldy, rule-laden, technical set of elements and definitions, especially within the short time frame that many workshops are restricted to. A summary of suggestions resulting from this focus group is available in the meeting results.

While all three educational environments are important, the focus of this project, being sponsored by the UCGIS, was on the university environment. Accordingly, a metadata education strategy and accompanying materials were developed based primarily on the university working group's recommendations, which are summarized in chapter 5 and in the meeting results. In addition, a review of existing university and college course outlines and materials and teaching methods for GIS-related courses was performed to assist in the synthesis of existing metadata training materials and workshop recommendations into a final product. Though the final web-based product, Education Strategies for Integrating Metadata is geared specifically toward metadata education in university courses such as "Geographic Information / Spatial Concepts", "Principles of GIS", and "Advanced GIS", we feel that the results are still applicable in other educational settings including technical colleges, professional development courses, workshops, and self-paced learning.

2. The importance of integrating metadata into GIS curriculum

Changes have been gradually taking place in GIS curriculum during the 1990s, as recognized by the formation of the University Consortium for Geographic Information Science (UCGIS) in 1994. The development of geographic information science (GISscience) as a discipline reflects the move beyond the mechanics of computer hardware and software application tools (traditional GIS) into a new, transdisciplinary research field. GISscience seeks to redefine geographic concepts and their use in the context of geographic information and, more broadly, the digital age (NCGIA, 1995). Forer and Unwin (1999) suggest that GISscience encompasses not only the technical and conceptual underpinnings of the use of geographic data, but the considerable social, legal, and ethical issues where are arguably of greater importance and equal complexity:

GIS are rapidly evolving to become standard tools, influencing everyday decision-making and acquiring the potential to penetrate even such market niches as the primary school classroom and home computer system. Mass-marketing of data goes hand-in-hand with this revolution. What are the ramifications of this penetration of GIS and spatial information into society? How many users of GIS will be appropriately informed users? GIS education must expand accordingly to accommodate this shift and a new balance will be defined between technical, conceptual, and societal issues. Technology is ceasing to block progress in learning GIS, allowing more opportunity to focus on depth and breadth, focusing on growing conceptual issues needed by such everyday usage (Forer and Unwin, 1999)

What are these conceptual issues that are needed to address the penetration of GIS into new markets and into everyday usage? Forer and Unwin (1999) list data quality, error propagation and fuzzy objects, in addition to social, legal and ethical issues. Marble (1998) includes basic spatial concepts such as projections and scale in addition to data quality and uncertainty. Considerable study and review went into the development of a curriculum for GIS education in Europe, in which developers noted that some course topics should be upgraded in importance in the new curriculum, including data upgrading methods, database design methodologies, basic steps in completion of a small application, and metadata. (Kemp and Frank. 1996).

As GIS technology has become increasingly "user-friendly" there has been the development of an erroneous notion that it can now be mastered by almost anyone with minimal effort. In actuality, this results in students who are only able to "apply perhaps 10 percent of the power of GIS technology, and this often incorrectly" (Marble 1998) because of the lack of a good foundation in basic spatial concepts. For instance, out of 12 GIS-related classes offered by the Community College of South Nevada, only one class has an instance of a learning outcome related to data quality, projections, or scale. In addition, there is also a growing trend among universities, research centers, and continuing education centers associated with universities to offer professional "short courses", 2 to 3 days in length. These courses provide an introduction to GIS and to specific GIS software for professionals seeking skills in this new technology. Within 2-3 days the basics of using the software can be covered, but in such a concentrated course very little time is allowed, if any at all, to cover basic issues relating to projections, scale, data quality and important societal and institutional issues related to GIS use.

Interestingly enough, the foundational basic elements that Marble identifies for a GIS curriculum are all topics that lead directly or indirectly to the elements of the content standard for metadata. In addition, many of the social, legal, and ethical issues of GIS also tie into importance of metadata as a means of cataloging GIS data and a measure of the fitness-of-use of data for particular GIS applications. Educational units on these topics opens the door to "weave" metadata into curriculum, especially through exercises involving downloading, converting, transforming, and using different data sources together for the sake of analysis.

Outside of changes in GIS curriculum, there are also trends in information technology/information science that highlight the importance of metadata. Tremendous growth in markets for data warehousing/cataloging and decision support systems has resulted in greater awareness of the need for metadata to deal with all the complexity inherent in collecting, organizing and transforming data into information for daily consumption and decision making (DBMS 1996). Metadata may be used in at least three different contexts: catalogues, management records, and accompanying data sets (Guptill 1999). As a result, it is a crucial, though often invisible, element to a wide range of operating environments, from the development and use of routine applications to complex decision support environments. For instance, metadata for web-pages are invisible to the majority of world-wide-web users, but are becoming increasingly important for providing meaningful results from web search engines. Veen (1997) suggests a future where standardized meta tags will enable search engines to produce results from resources around the world for advanced queries such as "research on limnology of polluted waters hosted by US universities and updated since January '97 – or entertainment magazines with articles about Beck prior to July '96 that aren't talking about Jeff Beck – or mailing lists that discuss dual-citizenship issues." Since any home computer user is capable of creating web-pages, metadata must become an educational issue both inside and outside of GIS realm in order to realize this future.

In the different but equally important realm of decision support, data analysts and executives are looking for useful facts and correlations that they may recognize only when they find them. This requires them "to get in among the data," and they need to be able to readily understand its structure and meaning in order to successfully accomplish this (ComputerWire Inc 1996). Metadata is key to making data more readily useable to decision makers. This is a good argument for integrating metadata awareness not only into GIS curriculum but into the broader arena of information technology and science curriculum.

Up until recently, metadata training has been conducted in workshops unrelated to standard curriculum. Workshop material often focuses on exhaustive coverage of metadata content and use of software for creating metadata within a short, concentrated time period ranging from less than a day to a couple days. Learning about metadata through this approach is tedious and likely to be soon forgotten as not all elements of metadata are used all the time. Various other obstacles were encountered in the process of metadata implementation in a program conducted by MetroGIS within the seven-county Minneapolis-St. Paul metropolitan area in 1998, one the first thorough studies of metadata implementation undertaken. This program provided a metadata implementation package including software, instructions, workshops and on-site visits to over 30 private and governmental organizations within the region, and included a follow-up plan to assess results. The participating organizations all recognized the value of metadata, or at least the potential; but implementation was a stumbling block for a high percentage of the organizations. The major reasons for failure to implement metadata was a reluctance to commit time to creating metadata because it would take too much time away from more important or necessary endeavors/responsibilities (Gelbman and Mathys, 1999). Organizations also expressed concern over the complexity of the federally-endorsed metadata content standard, even though they were introduced to a much abbreviated and simplified version of the standard developed specifically for Minnesota. The need for documentation had already been met by some of these organizations using their own internal format, and data-sharing was not a great enough necessity to provide motivation to change the documentation to a standard format. As a result of difficulties encountered in metadata implementation, Gelbman and Mathys (1999) recommend both short-term and long-term strategies in order to increase effectiveness. Short-term strategies were limited to providing support to organizations that are interesting in producing metadata but do not believe they have the resources to devote to it, and to continue marketing and peer-pressure on other organizations that are not yet interested or do not want to change from their own non-standard documentation system. Most importantly, Gelbman and Mathys (1999) stress that "any long-term approach to metadata implementation requires cooperation with educational institutions. Agencies with data access issues and metadata must establish contact with the respective departments at universities, colleges and technical schools to discuss measures to incorporate metadata into classroom material." Integrating metadata throughout a GIS curriculum provides more opportunity to tie in metadata with other important GIS issues and concepts, and to minimize information overload so often associated with learning to implement a complex standard.

GIS curriculum can also benefit metadata education based its experience in balancing technical skills in software use with necessary background knowledge of concepts and theory. "We must cease confusing mastery of software commands with attaining a grasp of intellectual concepts" (Marble, 1998). Some may argue that there isn't anything "intellectual" about metadata. Yet the content of metadata and the implications of standardized metadata use both tie in directly to the conceptual foundations of spatial information and the implications of its use within society.

3. Issues involved in curriculum development and pedagogy

As the importance of teaching metadata has become more widely recognized, educational issues have arisen that have already been uncovered and debated within the field of GIS education and curriculum development. Curriculum development concerns for metadata include such questions as these (Unwin 1997):

Before tackling these important questions, it is beneficial to look at some issues that have already arisen and been addressed specific to GIS education, which contain some relation to metadata education. Two issues in particular have surfaced and resurfaced quite often in GIS education literature: education vs. training and breadth vs. depth. Both GIS and metadata share some similarities, such as a profusion of technical terms and concepts, and a prevalence of software tools to perform basic functions for creation, management, and use of either GIS data or metadata. These similarities lead to difficulties in teaching both GIS and metadata. It is important to find a balance between education (with concepts and principles) and training (acquiring skills for operating a specific software) (Piscedda 1994). One approach is theory-driven and lacks skills training; the other is technology driven and lacks theory. A similar example of this dichotomy of approaches exists for traditional statistics courses, where the direction of common wisdom has turned to an emphasis on theory. A good basis in theory is essential and more important than learning to operate statistics packages (Kemp 1992). Kemp (1992) and Rogerson (1992) suggest that shortcomings in skills training can be overcome by developing short courses and workshops which provide intensive training in specific hardware and software. Their ideas are being affirmed as a growing number of software-specific short courses being offered by different institutions, many affiliated with universities. These short courses free up time in regular classes to be devoted to basic concepts and advanced issues. Since software trends are rapidly changing and unpredictable, the software used in last year's courses may be out-dated, no longer supported, or radically revised. Instead of having to revise class exercises to keep up with changing software trends, university courses can use "basic" software to teach GIS principles, while relying on technical courses to provide students with up-to-date skills in current software. This is also true with software for metadata creation. Since the adoption of the FGDC's content standard for metadata, there has been an explosion of software tools for creating metadata. Just as quickly as new software appeared, existing systems became out-of-date or no longer supported. Furthermore, mastery of software does not necessarily equate with mastery of a subject. Incorrect or inappropriate metadata can still be created even with the most user-friendly, powerful software, if users do not have a strong foundation in spatial concepts, and a good grasp of the implications of improper use of data.

Studies on the introduction of new technologies within institutions also offer some insight into the education vs. training issue. Introduction of new technology is not just a technical matter (Campbell 1999). Dale (1991) notes that in introducing new technologies or new methodologies, it is the ability to overcome the management and institutional problems, not the technical problems, that determines the success of a system. Training skills alone are not enough; potential users must understand how the skills relate to the larger institutional concerns. Information is an organizational resource, in addition to the basic resources of land, capital and labor. It is essential for a manager to understand the qualities of information and how it is used in decision making. These qualities include user needs, accuracy and precision, and acceptable risk (Dale 1991): all of which are concepts underlying metadata. In another business model, Heywood and Petch (1991) break down issues relating to adoption of GIS technology in a business into five categories:

    1. Conviction of the effectiveness: ease of use/transition; improvement of quality and quantity of productivity
    2. Movitation to implement the technology and learn how to use, apply, improve and develop it
    3. Skills necessary for computing: map design and construction, translating tasks into GIS functionality; interrogation and interpretation of spatial information; decision making with new types of information… new terminology, new structure
    4. Knowledge of spatial systems and their behavior; geographic characteristics associated with social, economic and environmental information and their mutual interaction; of the errors which are intrinsic to spatial data and influence their use
    5. Experience in thinking problems through in a spatial context; using the new tools associated with a GIS approach; problems which may be encountered as a result of adopting these new practices

According to Heywood and Petch (1991), the management and education problem is to govern the complex relationships between these parameters. For instance, educating/training for skills alone does not ensure effective implementation, since metadata, like GIS itself, is more than a software program: it is a whole philosophy of how to approach tasks. As GIS is implemented within an organization, all levels of that organization should be exposed to some degree of awareness as to why a spatial approach is required; similarly, all levels of the organization should be exposed to some degree as to why a standardized documentation approach is required. Toppen (1991) identifies educational goals for different organizational levels: a general manager; a GIS data or project manager; an "ad-hoc" user who uses GIS only occasionally; and a full-fledged analyst who uses GIS on daily basis. The common education goal of all these levels is a knowledge of basic concepts and a broad exposure to application/implementation issues. Similarly, students with different professional goals relating to GIS still require a basic understanding of spatial concepts and awareness of larger institutional/societal implications. Some students may choose to add specific skills to their foundational knowledge, but this training is dependent on their specific goals.

Given the need for metadata education, versus training, the next issue that arises in education is how to teach both depth and breadth. Within GIS curriculum, breadth can include scientific, societal problems, managerial, legal and ethic questions arising from use of GIS. Depth can include concepts from database management, computer programming, data models, inherent strengths/weaknesses of different systems, structures and algorithms. If metadata is to be integrated within GIS curriculum, opportunities must be explored to expand beyond basic awareness to include breadth (its relationship to standards, digital libraries, GIS implementation, decision-support) and depth (its relationship to database principles, quantification and communication of error/uncertainty and fitness-of-use). Certainly, including as much breadth and depth as possible is beneficial, but there is only so much information that can be taught without taking away time from other important GIS elements. Is there sufficient time and appropriate structure in introductory level GIS classes to provide breadth and depth? Integration of metadata within the elements of breadth and depth related to GIS is the key to enhancing existing educational content without actually having to add or substitute a great deal of new information.

The issue of what is and what is not appropriate educational content for different types of courses and different types of students is tied to a related pedagogical issue: content-centered versus student-centered curriculum models. Educational research has shown that content is not the major influence in student learning. The content that is intended to be taught filters down to what is actually taught, to what the students actually write down, to what is absorbed after modification by the students' additional work and interaction with others, to what is actually remembered and reproduced. The result is all largely dependent on individual student goals and learning processes (Unwin 1997). Another problem with content is that it becomes out-of-date quickly, especially in such rapidly evolving fields as GIScience and metadata. Student centered design is based on examining individual student needs and attempting to provide curricula to meet them. Ideally, this approach would allow an instructor and a student to interactively develop a course of learning together, defining specific, customized learning objectives. In reality, this is not yet possible, although new instructional delivery methods (self-paced, web-based) are making this more of a possibility. Considerable work has already been accomplished in developing flexible curriculum, materials, teaching methods and tools to assist in student-centered design or at least more flexibility for instructors in course design. These include:

These examples range from loosely-structured course resources to tools for designing courses to detailed, self-paced tutorials for learning specific topics. For instance, the NCGIA Core Curriculum in GIScience provides course materials which can be adopted for lecture format, but does not impose any specific structure or educational objectives, nor imply required content for GIS courses. Instructors are encouraged to pick and choose among the materials offered in order to develop courses suited specifically for their own students. Materials exist to provide both breadth and depth, but they impose neither upon any course structure. The NCGIA has another project, the Core Curriculum for Technical Programs (CCTP), which is task-oriented and focuses on how to use the technology effectively. Rather than addressing topics such as error from an abstract perspective, the CCTP, for example, provides materials for the instruction of digitizing which includes a tangible demonstration of the relevance of those aspects of database error which arise during the digitizing process. The CCTP is also more structured than the CC for GIScience, as learning topics are organized into sections designed to educate students to a desired level: awareness, competency or mastery. Topics are also provided with links to preparatory and complementary topics, which allows non-sequential movement between topics in addition to the basic sequence implied by the outline of topics.

This flexible directional organization is a technique used by several tutorial programs developed to teach specific topics. Thompson (1991) discusses how creative learning opportunities and student-centered learning can be manifested through the use of hypermedia concepts: "information provided in an object-oriented, user-controlled, dynamic approach within a loosely structured, discovery based learning environment." The idea in this approach is to let the student pick their own way through topics without imposing a structure upon them. Raper and Green (1992) developed a hypermedia tutorial based on these same premises, with associative links between concepts, implementation, and applications to help students learn about concepts in a contextual environment. One advantage with hypermedia tutorials is the ability to spread out a large amount of information across many discrete places without creating redundancy. Another advantage is the ease of dynamic "scale change": providing the ability to explore a topic in more depth or to explorerelated topics for more breadth (Thompson 1991). Disadvantages may include disorientation for a novice user browsing a large database with relatively little structure, though links to index maps and tables of contents can help provide an overview of structure. Also, while hypermedia tutorials may provide a more student-centered environment, they do not necessarily increase students’ retention. More effective learning strategies call upon students to produce rather than merely consume information (DiBiase 1996). Lessons learned from the GeographyCal project (Healey 1998) stress the importance of developing active learning experiences. Active learning principles that require student involvement, discussion, decision making and problem solving have proven to be very successful (Keys-Matthews 1999). Laboratory work is also an important method of active inquiry. "Where lectures and readings at best familiarize students with concepts, exercises and discussions provide opportunities for students to develop some level of mastery by engaging the concepts experimentally" (DiBiase 1996, Boud et. al 1986). This premise is upheld by the Coulson and Waters (1991) review of the NCGIA core curriculum, which noted that the most successful lectures were those which use the case study format, applying concepts in real-world scenarios by evaluating their strengths and weakness.

An international workshop, Interoperability for Distributed GIScience Education, was held in 1998 to discuss these important educational issues (Kemp et al 1998):

Several educational approaches were reviewed as potential methods for addressing the above issues, the basic theme being that educational materials need to become "interoperable". Like separate modules of software that can interact with each other, interoperable education materials coming from different sources and authors can be used to put together courses and to restructure courses as necessary. The modular component is important not only because of the need to update materials to keep pace with changing technologies, but also to allow different disciplines to define their own emphases on the subject matter or to have the ability to take different perspectives on the same subject matter (Langford et. al 1994). Links to modules in related subject networks are important, because many of the ideas discussed are transferable (Healey 1998). With interoperability, the modules can form a cohesive whole and allow for a meaningful construction of course materials from different sources. The Instructional Management Systems Project (IMS) (www.imsproject.org) represents a consortium of government, academic and commercial organizations who are developing a set of specifications and prototype software for facilitating the growth and viability of distributed learning on the Internet, based on the above principles. Apart from geospatial data, metadata is an issue to the development of distributed learning on the internet, since web-based education materials need to carry their own metadata in order for projects such as IMS to be successful. Some of the concerns voiced by participants at this workshop were: given the need for localization in geographic information science, how should learning profiles or educational settings be matched to metadata? and, can we establish hierarchical schemas in metadata to address geographical or disciplinary foci? It is ironic that some of the impetus for metadata education may come from the need for metadata in educational material itself!

ESRI’s Virtual Campus (campus.esri.com) was one of the educational approaches reviewed at the workshop. The Virtual Campus is of particular interest because this program has developed a "Knowledge Base" to assist in course development. It is essentially a database of GIS concepts, examples, exercises, and test questions that can be retrieved and structured according to the wishes of a course author. Additionally, the Knowledge Base can be used to publish the content of a course as a web-based training module, as self-study workbook, or as notes for a lecture-based course. According to the workshop report (Kemp et. al 1998), "the Knowledge Base takes the opportunities for interoperability in GIS education to a new level. The database of GIS concepts, examples, and exercises, structured into bite-sized components ready for re-combination to meet the needs of a particular tutor, seems to offer an attractive model for the management and use of interoperable course content." The Knowledge Base model is also applicable for integrating metadata education within GIScience. A knowledge base of metadata concepts, examples and exercises can be developed as components which GIScience course authors can pick and choose from to integrate into their course. A knowledge base is more than a database or a clearinghouse of existing materials, it also provides a framework and sometimes even a toolbox for constructing courses as well as techniques for how to teach certain topics based on expert knowledge or experience.

A final note on curriculum development and pedagogy issues is Jenkins' (1991) statement that "there is always an immense concern with what should be taught while the how of teaching is seldom carefully analyzed". While this statement was directed toward GIS education, it also applies to metadata education. Past metadata education has focused almost entirely on content. Lessons learned from the GIS arena of education should be that content cannot be successfully learned and integrated into standard professional practice without careful study into teaching methods, not just content. According to Jenkins (1991), a basic model of curriculum is an interaction between aims and objectives, methods of assessment, teaching methods and content. Changing one element will affect the others and what the student learns. It is important to define aims for long-term goals as to the purpose of education and objectives for immediate targets for what one expects students to know or do as a result of a particular course. Defining learning aims and objectives, evaluating different teaching methods, and following up with methods of assessment, all contribute to developing a less subject-centered and more learning-centered view.

4. Review of GIS-related courses: course content and materials

In order to provide a structure and a "knowledge base" for integrating metadata into GIS-related course work, it was important to survey existing GIS-related course content and materials to evaluate the differences in content and emphasis of topics in a range of courses. In addition, the survey was used to study course topics in more detail to evaluate which topics are conducive for integrating metadata and specific methods for integrating metadata material. The starting place for this survey was existing course materials such as the NCGIA's Core Curriculum for Geographic Information Science (http://www.ncgia.ucsb.edu/education/curricula/giscc/) and Core Curriculum for Technical Programs (http://www.ncgia.ucsb.edu/cctp), and University of Texas at Austin’s materials for the Geographer's Craft (http://www.utexas.edu/depts/grg/gcraft/contents.html). In addition, links to other on-line course materials are available through University of Texas at Austin's Virtual Geography Resources (http://www.utexas.edu/depts/grg/virtdept/resources/educatio/courses/gis/gis.htm) and many courses are also available through participating departments in member institutions of UCGIS (http://www.ucgis.org/fMembers.html). A three volume publication from 1992 (Aangeenburg) exists containing curricula, course outlines and laboratory exercises of GIS courses from six different disciplines, but given the rapid evolution of GIS, GIScience and associated software and corresponding shifts in education, it was determined that more recent materials available on-line would provide a more up-to-date source for this aspect of the project.

Some limitations regarding the survey should be noted. First and most importantly, only course information and materials available on-line were included in this survey, as the project time-frame did not have resources for contacting individual departments or professors to request hard copy materials. In many cases this meant that information was incomplete or possibly not up-to-date. For instance, some departments or instructors provided almost all their materials on-line and required students to participate on-line as well; for other courses, only course syallabi and outlines were available on-line; in rare cases only brief course description were available. The survey was also limited in that it was not a comprehensive or structured sample of existing GIS-related courses. There are undoubtedly many other course outlines and materials available for other institutions who are not UCGIS members or included in the University of Texas' resource list. However, this survey did achieve a broad overview of many different course types, including traditional university courses, GIS certificate courses, distance-learning courses, community college courses, and professional (short) courses.

Elements that were included for all courses in the survey were the course type, course name, institution and department. Defining course type categories was important in order to develop different strategies for integrating metadata. Course outlines (and materials, if they were available) were reviewed in the survey to determine whether topics such as metadata, data quality issues, data sources, GIS project design or implementation, standards, legal or ethical issues, decision making or decision support system, and GIS trends/GIS in society were included in the course. All of these topics were included in the survey since they each relate to metadata in some aspect and can be used to integrate metadata education into a course. Other metadata-related topics, such as projections/scale, database principles (attribute data), data models and spatial analysis were not specifically included in the survey since these topics are basic to almost all GIS courses, while the previous list of topics are more "optional". Generally only 2 to 3 of these topics appear in any given course, so it was useful to determine in which type of courses these topics were most likely to appear. The survey also kept track of courses which also required laboratory (hands-on) exercises and student projects. Any available materials (lecture notes, exam reviews/sample questions, lab exercises, project descriptions) were reviewed in order to identify existing or potential methods for integrating metadata.

A total of 145 courses were reviewed in this survey. The majority of these courses only had course outlines and syllabi available on-line for review; 34% also had on-line lecture materials availabe (often in the form of lecture outlines or Powerpoint slides) and 29% of the courses had labratory exercises available. The courses were classified into 8 broad categories of course types. Of these, the most common course type was a basic "Introduction to Geographic Information Systems" course (46 examples). GIS applications/ issues courses were the next most common (24 examples), followed by "spatial awareness" courses (21 examples). All course types are listed in Tables 1 and 2.

Table 1. Total number of surveyed courses for each course type (y-axis) and number of courses including each course type (x-axis) within their outlines or material. Explanation of field names
  Total
Number
Metadata Data Quality Data Sources Standards Implemen-
tation
Decision making Future On-line
Spatial awareness 21 8 11 17 5 10 4 11 9
Intro GIS 46 23 32 30 8 20 9 19 21
Advanced: applications or issues 24 5 12 11 4 13 8 4 11
Advanced: software specific 9 0 1 2 0 3 2 0 0
Advanced: spatial analysis 11 2 4 5 0 1 2 2 4
Other disciplines teaching GIS 17 5 10 13 2 7 4 2 3
Cartography 5 0 3 5 0 0 0 1 1
Short courses 12 5 3 7 1 6 0 0 1

Table 2.Total number of surveyed courses for each course type (y-axis) and percentage of courses including each course type (x-axis) within their outlines or material. Explanation of field names
  Total
Number
Metadata Data Quality Data Sources Standards Implemen-
tation
Decision making Future On-line
Spatial awareness 21 38 52 81 24 48 19 52 43
Intro GIS 46 50 70 65 17 43 20 41 46
Advanced: applications or issues 24 21 50 46 17 54 33 17 46
Advanced: software specific 9 0 11 22 0 33 22 0 0
Advanced: spatial analysis 11 18 36 45 0 9 18 18 36
Other disciplines teaching GIS 17 29 59 76 12 41 24 12 18
Cartography 5 0 60 100 0 0 0 20 20
Short courses 12 42 25 58 8 50 0 0 8

The first of the course types identified was the most introductory, providing students with "spatial awareness" about geographic data and related technologies; this type of course may be loosely labeled as "Intro to Geographic Information" or "Spatial Concepts". This is a foundational class for any GIS, GIScience or Cartography curriculum, since it covers basic topics that most Cartography, GIS, GPS, and remote sensing courses have in common. It also provides spatial awareness for undeclared freshmen/sophomores or professionals coming from other disciplines interested in GIS. Penn State permits this course to be used as one of the university general requirements for social sciences. This different between this course and a Cartography course is that it focuses on map concepts and maps use rather than actual map production. This type of course includes several topics that are conducive to the integration of metadata: principally the discussion of projections and scale, and the variety of different types of geographic information available.

Even more potential topics for metadata were realized in the next course type, "Introduction to GIS" or "Principles of GIS". This is the most common type of GIS-related course offered in higher education, and the most comprehensive in the breadth of topics covered. This type of course can include everything from basics such as projections and scale on up to data models/structures and spatial analysis, and may touch on a diversity of topics from GIS project design/implementation, GIS in society and ethical or legal issues related to use of GIS.

There were several variations of "Advanced GIS" courses, depending on whether the course was designed to teach specific software skills, to cover advanced issues, to provide an overview of GIS applications, or to cover spatial analysis techniques and algorithms in more detail. Though somewhat more limited than the "Introduction to GIS" course, because the focus is more specific, there still exists two or three potential topics conducive for integrating metadata. It is also possible that in such advanced courses, focus may be placed on one or more GIS implementation issues that can include a more in-depth look at metadata. For instance, an advanced "GIS Workshop" course offered by the University of Washington includes learning and using the metadata content standard as one of the main learning objectives of the entire course.

Other types identified were courses from other disciplines (e.g. Forestry, Planning, Landscape Architecture, Geology) introducing GIS or including GIS in their content; Cartography courses, and professional (short) courses either introducing GIS concepts or focusing on software training. Remote sensing courses were not specifically included as part of the survey, though two were included in the survey (one at University of Idaho and one short course offered by Berkeley). In addition, attention was given to the University of Maryland's Remote Sensing Core Curriculum. Global Positioning Systems (GPS) is typically covered as topic within spatial awareness and applications courses, or dealt with specifically in short courses.

Of the 145 surveyed courses, 48 (33%) made explicit reference to metadata either within the course outline or within course materials (Tables 1 and 2). The number of courses including metadata is probably higher than this figure, as metadata could very well be included in the course without being explicitly included in the course outline. 50% of all surveyed Introduction to GIS courses (46 out of 145) included or mentioned metadata; 8 out 21 (38%) of spatial awareness courses; 5 out of 12 (42%) of short courses; 5 out of 17 (29%) of other discipline courses teaching GIS; 5 out of 24 (21%) of advanced issues/applications courses; and 2 out 11 (18%) of spatial analysis courses. Metadata was not mentioned at all in any of the 9 software-specific courses or cartography courses. In 22 of the 48 courses, metadata is only briefly mentioned as part of a lecture (most commonly a lecture on data sources or on databases), in a lab, or as an exam question ("describe what metadata is and five components of it"). In the other 26 classes, metadata was dealt with in more detail, usually within a labratory exercise or as part of a class discussion. Most of the laboratory exercises have to do with finding appropriate data and downloading it off the Web, requiring the ability to read and interpret metadata; however there were four instances of classes (one spatial awareness, one introduction to GIS, and two advanced classes) requiring students to create metadata, as well. In several courses, students are required to produce data documentation for their final project. Though the documentation is not explicitly refered to as metadata, data dictionaries and data quality reports fall within the definition of metadata.

The other course topics covered in the survey all have potential to make direct or indirect reference to metadata. Data Quality is a major component of a complete metadata report, and this is a topic that is dealt with in 70% of all Introduction to GIS courses, 52% of all spatial awareness courses, and 59% of all discipline courses teaching GIS, and 50% of all GIS applications/issues courses. Metadata is critical to finding and evaluating data for use: the topic of Data Sources is also frequently covered in GIS-related courses (81% of spatial awareness courses, 65% of Introduction to GIS classes and 100% of Cartography/mapping courses surveyed). Other topics, such as standards, ethics, implementation issues, future (GIS trends),and decision making (e.g. GIS and decision support systems) do not appear as often but still offer additional means to integrate metadata as well as to help students realize the extensive, far-reaching implications of metadata.

In reviewing the course outlines and material, the survey also kept a record of courses that make use of other on-line materials, such as the NCGIA Core Curriculum for GIScience or the Geographer's Craft topics. Of the three most common GIS classes (Intro to GIS, GIS Issues/Applications, Spatial Awareness), from 43 to 46% of the surveyed courses included references to on-line materials. On-line materials not only included material produced by universities but also private companies (ESRI's Virtual Campus and Conference proceedings) and governmental organizations (USGS). The Federal Geographic Data Committe (FGDC) web pages on metadata and the National Spatial Data Infrastructure were also referenced by several classes. Some classes used these web sites as well as specific GIS project web sites as a basis for discussion topics. Another common activity in courses was to require students to find and survey GIS-related web-sites, information which is often compiled into a list of resources for future students to utilize. The relatively frequent use of the World Wide Web by GIS-related courses, as a supplement or sometimes a complete replacement of traditional course texts, seems to uphold the points made by the 1998 interntational workshop, Interoperability for Distributed GIScience Education, (see Chapter 3). The highly dynamic nature of the GIS field, coupled with a changing model of higher learning education into a more flexible lifelong learning environment, both point toward increasing use of the WWW as a medium for GIS and integrated metadata education.

5. Synthesis of review and recommendations

An important result of the June 3-5 1999 meeting of metadata and GIS education specialists was a list of desired products or efforts that the three different working groups (university, self-paced/distance learning, and workshop) felt were important. Participants were asked to rank these products in categories of high, medium, and low priority. The main focus of this project, being sponsored by the UCGIS, was on the university environment, and the final product was developed according to the high priority recommendations of the university working group, synthesized with research on curriculum and pedagogical issues (chapter 3) and a review of existing course outlines and materials for GIS-related courses (chapter 4). Two of the university group's highest recommendations were in common with recommendations made by the other two groups, and are discussed at the conclusion of this paper. The other two recommendations of the university group included the development of exercises incorporating metadata aspects as a high priority, and several example course outlines indicating which topics in GIS classes are appropriate for referencing metadata.

Both of these recommendations are closely tied to a major theme identified by the meeting participants, that of "weaving" metadata into existing curriculum and education materials for the field of GIS. Metadata should not be treated as a separate component, but as an integrated part of the entire GIS process. The practice of presenting the metadata content standard sequentially, element by element, was informally referred by the meeting participants as "the death march". Such a teaching method entombs the topic of metadata into its own self-contained unit, or forces it into the realm of format-specific training – analogous to teaching specific GIS software commands instead of the generic concepts used by all software. One meeting participant noted that the content of metadata in this respect has some similarities to map projections: a vital element in proper use GIS, but involving so many technical details that students can lose sight of the forest for all the trees. By bringing up different aspects of metadata at different intervals throughout a course, the information overload of the "death march" can be minimized. According to this suggestion, the final product was designed as a collection of topics that are always or often covered in GIS-related courses, (e.g. projections and scale, data models, data sources, spatial analysis, GIS project design/implementation). Surveys of course outlines and materials were used extensively to determine which topics are used most often in which type of courses and how existing learning material for these topics may be modified to include a greater emphasis on metadata.

The first specific recommendation of the university working group was the development of exercises incorporating metadata aspects. A related suggestion from the meeting was that with similar technical topics like projections and metadata, background concepts are best introduced during lectures and then followed up with specifics during exercises. Exercises are particularly useful if structured to drive home the point that without metadata (or likewise without an understanding of projections) the process of collecting and analyzing GIS data can be frustrating and inefficient at the least, inaccurate or inappropriate at the worst. The consensus of meeting participants appeared to be that exercises are important instructional strategy for metadata, especially exercises structured to require students to use metadata within an active, decision-making or problem-solving context. "Students do not learn by just listening, taking notes and studying for hours… [we] learn when we analyze, examine, discuss and apply information" (Keys-Matthews 1998). Further advantages of active learning principles are identified by Meyers and Jones (1993): "when we involve students in activities that lead them to discuss, question, clarify, and write about course content, we not only foster better retention of subject matter but help expand students' thinking abilities as well." Specific suggestions for exercises can be viewed in the meeting results and were used in the learning material developed as a final product for this project. In addition, discussion questions for some topics are included to help facilitate active learning.

The review of existing course materials provided examples of exercises which either already incorporate metadata or which are amenable for including metadata. Kemp (1991) notes that generic exercises, offering sample questions, analysis, and applications, are preferable than specific exercises using prescribed software or data. Step-by-step instructions are too error prone and too difficult to modify, not to mention that software required by specific exercises is very prone to change, requiring the necessity of update. With the availability of so much data now on the internet, there is no longer the need to provide students with "canned" data sets and active learning is facilitated by requiring students to find, evaluate, download, import, transform and use data that is available through clearinghouses and on-line project sites. In fact, quite a few existing courses do include on-line exercises such as this. In addition, nearly half of the surveyed courses also required a final project, designed to mirror a "real-life" GIS project in simplified terms. Students are required to go through all or most of the steps of a GIS project, including creating, editing, transforming and bringing together all necessary data sets in order to perform some types of analysis. This type of "exercise" is well-suited both for metadata use and creation. Since methodology is an important aspect required of the final project (whether it is presented as a poster, class presentation or paper), metadata fits right in as a method for recording sources and processes used. Students are typically required to keep a log of their activities as they work through their project. With the addition of some background reading on the topic of metadata, students can identify their logs as a form of metadata, or actually convert the logs into more formal metadata. As the logs are used and added to throughout the project development, this becomes an effective form of integrating metadata into the education process.

The second specific recommendation of the university working group was several example course outlines indicating which topics in GIS classes are appropriate for referencing metadata. The survey of existing course outlines and materials, described bed in Chapter 4, was used to identify different categories of GIS-related courses and typical topics covered by these courses. In the final web-based product, Education Strategies for Integrating Metadata, a list of commonly used course topics for each of these course types is provided, along with a recommendation "high" " medium" or "low" priority for integrating metadata into each of these course topics. For instance, the results of the course survey indicated that the topic "Implementation" is covered in 54% of "issues/applications" courses, but is only covered in 9% of the classes in the "spatial analysis" category. Therefore, this topic ranks as "high" priority for integrating metadata in an issues/applications course, and "low" priority in a spatial analysis course. However, in a spatial analysis course the emphasis on metadata will be in the importance of documenting steps taken in spatial analysis (results of analysis must be replicable), or in other topics such as Data Quality or Data Sources.

The three working groups at the meeting also suggested that different learning objectives or outcomes should be identified for different audiences: the "novice", the "informed user", the "GIS analyst" and finally, the "GIS Programmer". The use of learning objectives is an important tool for assembling learning content with a focus on student needs. Rather than developing learning objectives based upon the above audience categories, or other categories (objectives for attaining awareness, competency and mastery used by the NCGIA Core Curriculum for Technical Programs), learning objective categories for this project were modified from the five components relating to the successful adoption of new technology in a business by Heywood and Petch (1991):

  1. Conviction of the effectiveness: ease of use/transition; improvement of quality and quantity of productivity
  2. Movitation to implement the technology and learn how to use, apply, improve and develop it
  3. Skills necessary for accomplishing various aspects the new technology, including new terminology and new structure
  4. Knowledge of topics to add breadth (relation of the technology to other technologies/fields, implications within the organizational or social structure) and depth (origin and basis for the technical/organizational structure; deconstruction of requirements)
  5. Experience which equates to skills plus knowledge: ability to find deficiencies or flaws in existing implementation; ability to adapt techniques to new or unique situations; ability to provide alternative strategies; ability to develop new applications and methods

Learning objectives for each of the topics were provided for Conviction, Motivation, Skills and Knowledge (Experience being the outcome of these first four). These four categories of learning objectives were selected as the best representation of differing needs of audiences, since the meeting participants noted that potential audiences can range on a continuum in several aspects:

The five parameters noted by Heywood and Petch (1991) are best suited for this broad range of needs because it closely mirrors the complexity of the management and education problems faced in introducing a new "way of doing things" within an organization with already established procedures. As mentioned earlier, effective implementation is more than just training for skills, or providing access to necessary knowledge, since metadata, like GIS itself, is more than a software program but is a whole philosophy of how to approach tasks. A person with a technical position in an organization may select learning objectives to a level either of awareness, competency, or mastery. However, a manager who must decide whether or not the people in his/her division should take time away from existing tasks to implement a new task needs more than just awareness or competency; he/she needs material to provide conviction and motivation, and in addition enough knowledge to understand the implications and issues involved in the implementation of new tasks.

The final web-based product is an incorporation of these three elements: learning material and example exercises/discussion questions provided to help integrate metadata into various course topics; example course outlines provided for different types of GIS-related courses; and learning objectives organized into categories for educators and/or students to choose based on a desire/need for conviction, motivation, skills or knowledge. Accordingly, the web-based material is organized into three sections: education strategies by course topic, course type and learning objective. A prospective educator and/or student can approach the material from any of these three starting points. Underlying this structure, the material is interlinked so that all material is available from each starting point, though presented in a different format. For instance, within each course topic different learning objectives are listed, from which an educator or student choose to focus on. Alternatively, if the educator or student decides to start off by selecting specific learning objectives, these objectives are linked to related course topics which provide the material for accomplishing the desired objectives. The third approach, by course type, provides a table that organizes the learning material by course topic (y-axis) and the four learning objectives categories (x-axis). In addition, a ranking of "high", "medium" or "low" is suggested for the material, based on how appropriate it is for integrating metadata for the particular type of course.

The World Wide Web was chosen as a medium for providing these educational strategies and materials for several reasons. The main advantages of web for delivering educational resources is the ability to continually revise and update materials, as well as to provide links to other existing materials instead of being forced to produce redundant material. DiBiase (1996) also notes that the Web is an excellent medium for accomodating asynchronous group activities: discussion forums or group projects for distance-learning students. This type of education is becoming more important, as Kemp et. al (1998) notes there is a shift from a traditional, one-time-through university education experience to a flexible lifelong learning environment and an increasing demand for effective and efficient learning opportunities. The Web also provides a workable solution for the need of "interoperable" educational materials, coming from different sources and authors, as long as some elementary structure (metadata!) is provided so that different materials can be used together in courses and restructured as necessary for different student needs or changing trends. The educational strategies developed by this project were organized with the idea of providing a knowledge base: not just a database of different exercises and lecture materials and resources that are useful, but a knowledge base with organized links between topics and associated learning objectives. With a knowledge base, users of the material are given direction so that they can choose whether to go broader or deeper on a particular topic, or traverse from skills to knowledge or knowledge to skills, in order to achieve desired learning outcomes.

As a final note, it is important to mention the two highest priority recommendations identified by all three working groups at the 1999 meeting. These desired products were a marketing strategy to promote metadata and a revised and updated version of the FGDC's content standards workbook (the "Green Book"). Certainly both the results of the meeting and the background research both point to the importance of better marketing strategies for metadata, both within educational and professional circles.

Metadata has acquired a "bad taste in the mouths of GIS gourmets, in a manner analogous to garlic. Raw garlic tastes awful, but the professionals espouse garlic's health benefits. Metadata has been the topic among professionals for some time, but now the time has come to find ways to make metadata more palatable to the GIS community" (Gelbman and Mathys, 1999).

The results of this project seem to indicate that one of the ways to make metadata a more palatable subject is not just to tout its importance (which has been the focus of most previous marketing efforts; see FGDC's two brochures, "The Value of Metadata" and "Geospatial Metadata") but to demonstrate its integral place both within GIS data itself and the larger GIS process. At the June 1999 meeting, a representative from a vendor of GIS software, ESRI Inc, described metadata as an integral component of GIS structure. Traditionally, GIS has been described has being composed of two types of information: information about spatial location and relationships (coordinates, topology) and attribute (descriptive) information about geographic features or locations. A new potential marketing strategy would be to present GIS as being composed of three types of information: spatial, attribute, and metadata. The usefulness of this marketing approach is uphelp by a recent trend in GIS software to automatically "read" metadata, in addition to the traditional ability to read spatial and attribute information. For instance, some software can now interpret metadata in order to perform overlays between data sets in different projections, and other software can create metadata automatically as data is created or recombined/transformed within the software. The trend will hopefully continue that future versions of GIS software will also be able to automatically inform users of liability statements or use restrictions as they access data sets, and to better communicate/visualize data quality to users.

Part of the new marketing strategy to present metadata as an integral component of GIS data and the GIS process is to place less emphasis on the format of metadata and more on the content of it. For the average user and producer, metadata content is the most important thing: you need to know about the details of your data set, not the details of the format that information is stored in. Details relating to the specific format of the content are certainly helpful, but not necessary. Formats change rapidly like every other technology does these days, and it can be a drain on time trying to keep up with them. Content may change slightly over time, but not enough to be a major concern, just as the advent of computer-cataloging did not drastically change the content of a library card catalog, only the format of it. Unfortunately, too much attention in the past has been focused on the format of metadata content, which has caused much frustration and lack of motivation to use or create metadata. For this reason, a revision of the existing workbook for the metadata content standard (the "Green Book"), though certainly a helpful step, should not be considered as high a priority as products that help data users and producers discover the usefulness of metadata within their occasional or day-to-day forays into the world of geographic information. Since the focus of this project has been to provide educational strategies, the next step might be to revise and enhance the Green Book so that it can be used as a reference along with these suggested educational strategies. Used alone, the Green Book can be frustrating to those new to metadata, since it places more emphasis on the structure of metadata instead of the actual content. A shift of focus back to the basic content can reverse the tendency towards frustration and lack of motivation to use or create metadata, and at the same time help educate data users and producers about important spatial concepts in the rapidly expanding field of spatial technologies such as GIS, GPS and remote sensing.

 

 

 

Literature Cited

Aangeenbrug, R. 1992. Curricula, Course Outlines, and Laboratory Exercises. International Journal of Geographic Information Systems 6 (4): 337-340.

Boud, D. Dunn, J and Hegarty-Hazel, E. 1986. Teaching in Laboratories. Guildford: Society for Research into Higher Education, NFER/Nelson

Campbell, HJ 1999. Institutional consequences of the use of GIS. Geographic Information Systems Vol 2 Management Issues and Applications 2nd Ed. Longley, PA, MF Goodchild, DJ Maguire, DW. Rhind. Editors. John Wiley and Sons, NY.

ComputerWire Inc 1996. Data Warehousing Tools Bulletin: Briefing Paper: What is Metadata. http://www.computerwire.com/bulletinsuk/212e_1a6.htm (Article: March 1 1996, accessed Dec 1 1999)

Coulson, M.R.C. and N.M. Waters. 1991. Teaching the NCGIA Curriculum in practice: assessment and evaluation. Cartographica 28(3): 94-102

Dale, P. 1991. Education in Land Information Management. Cartographica 28(3): 23-30

DiBiase, D. 1996. Rethinking Laboratory Education for an Introductory Course on Geographic Information. Cartographica 33(4): 61-71.

Forer, P and D. Unwin. 1999. Enabling progress in GIS and education. Geographic Information Systems Vol 2 Management Issues and Applications 2nd Ed. Longley, PA, MF Goodchild, DJ Maguire, DW. Rhind. Editors. John Wiley and Sons, NY.

Gelbman, R. and T. Mathys. 1999. Making Metadata Part of Your Daily Diet. Proceedings of the 1999 Urban and Regional Information Systems Association (URISA) annual conference.

Guptill, S.C. 1999. Metadata and data catalogues. Geographic Information Systems Vol 2 Management Issues and Applications 2nd Ed. Longley, PA, MF Goodchild, DJ Maguire, DW. Rhind. Editors. John Wiley and Sons, NY.

Healey, M. 1998. Developing and disseminating good educational practices: lessons from geography in higher education.The International Consortium for Educational Development in Higher Education's Second International Conference on 'Supporting Educational, Faculty & TA Development within Departments and Disciplines', Austin, Texas, 19-22 April 1998

Heywood, D.I and J.R. Petch. 1991. GIS Education: A Business Perspective. Cartographica 28(3): 10-22.

Jenkins, A. 1991. Through a model darkly: an educational postscript. Cartographica 28(3): 103-108.

Kemp, K., D.E. Reeve, I Heywood, 1998. Report of the International Workshop on Interoperability for GIScience Education (IGE '98). Soesterberg, The Netherlands May 18-20, 1998. http://www.ncgia.ucsb.edu/ige98/report/

Kemp, K., A. Frank. 1996. Toward a consensus on a European GIS Curriculum: the international post-graduate course on GIS. International Journal of Geographic Information Systems 10 (4): 47-497.

Kemp, K. 1991. The NCGIA Core Curriculum Evaluation Program: a Review and Assessment. Cartographica 28(3): 88-93.

Keys-Matthews, L. 1998. Active Learning Pedagogy for the GIS ACCESS Project at Cypress College: http://gisaccess.cypress.cc.ca.us/

Langford M., Dykes J.A. and Wood J.D. 1994. The Design and Implementation af a Flexible GIS Teaching Resource. Fifth European Conference and Exhibition on Geographical Information Systems.

Marble, D. 1998. Rebuilding the top of the pyramid. ESRI: ARC News, spring 1998. http://www.esri.com/news/arcnews/spring98articles/01-rebuilding.html

Meyers, C. and T.B. Jones, 1993. Promoting Active Learning: Strategies or the College Classroom. San Francisco: Jossey-Bass Publishers.

Nyerges, T. and N. Chrisman. 1989. A Framework for Model Curricula Development in Cartography. The Professional Geographer 41(3): 286-295.

Piscedda, S. 1994. How to set up a GIS course: the contribution of Formez through an Italian experience. Fifth European Conference and Exhibition on Geographical Information Systems.

Raper J. and Green N. (1992), Teaching the Principles of GIS: Lessons from the GISTutor Project, Int. J. GIS 6(4). 279-290.

Rogerson, R. J. 1992. Teaching generic GIS using commerical software. International Journal of Geographic Information Systems 6(4): 321-331.

Thompson, D. 1991. GIS A View from the other (dark?) side: the perspective of an instructor of introductory geography courses at the university level. Cartographica 28(3): 55-64.

Toppen, F.J.1991. GIS Education in the Netherlands: A bit about everything and everything about a bit? Cartographica 28(3):1-9

Veen, J. 1997. XML: Metadata For the Rest of Us (Part 1) by Jeffrey Veen. http://www.wired.com/news/news/technology/story/4997.html

Unwin, D.J. 1997. Unit 159 Curriculum Design for GIS. NCGIA Core Curriculum in Geographic Information Science. http://www.ncgia.ucsb.edu/education/curricula/giscc/units/u159/u159.html