Data Sharing Projects
The following final report of the year 2001 refers to five in 2001 existing projects and their websites. Unfortunately some projects do no longer maintain their website. Therefore some of our links are outdated. This is especially true for "Globe" and "Roadkill".
The following links are active in february 2010:
(1) Journey North: http://www.learner.org/jnorth/
(2) Estuary Net: http://yosemite.epa.gov/water/surfAH.nsf
(3) Water on the Web:
Data Sharing Projects -
A Critical Analysis of Five Projects from the Perspective of Competencies and Opportunities for Interactive and Exploratory Data Analysis
Draft Final Report, February 2001
Rolf Biehler & Stefan Schweynoch
University of Kassel, Dept.of Mathematics and Informatics
Heinrich-Plett-Str. 40, 34132 Kassel, Germany
(Institut für Didaktik der Mathematik IDM, Universität Bielefeld, until 1999)
1. Introduction
These reports were were written as part of a larger project, directed by Prof. Cliff Konold, University of Massachusetts at Amherst, on "A study of student investigations in data-sharing projects" (NSF Grant No. REC-9725228). For details about the background of this project and the intentions of these technical reports see background of the project (written by Cliff Konold).
2. Principles of the technical reports on the 5 data sharing projects
The technical reports have practically all the same structure, which we chose according to our intentions:
- A. Summary
- B. Introduction
- C. Data and data archives
- D. Tools for data analysis
- E. Data analysis in the curriculum
The Introduction provides a brief summary of the project so that our analyses that will follow can be understood and so that the reader gets an idea of the project. We use as many texts from the project's website as we can and show these quotations by using smaller fonts. We provide links to the original project site.
In Data and data archives we describe in detail which data (variables, measurement procedures, metadata), where on the web and in which format they are available. Projects differ very much in how well the data are structured and how well the definition of variables is documented. Sometimes a specific interface supports the selection of data from a large archive. We regard a data archive separated from the rest of the material but linked to it as advantegeous.
The paragraphs on Tools for data analysis analyzes which tools are provided by the projects to analyze their own data and for which data analytical purposes they are adequate. Relatively simple and problem-adapted on-line tools make it easy for students to do data analysis. Often, however, the limitations due to the simplicity are severe in some cases. If data are also available in a standard format that can be imported into a data analysis software or a spreadsheet this is a valuable supplement.
The core of our analysis can be found in Data analysis in the curriculum where we first of all analyze the role of data analysis in the current project and which type and support is provided for data analytical activities. The supports for data analysis that we look for are software tools, subject matter knowledge for data interpretation, data analytical and statistical knowledge and strategies, exemplary data analysis and expected answers. In general, we found that most projects could improve their sites if they added data analytical and statistical knowledge and strategies to their material. Project authors seem to underestimate the need for this and often seem to think that process goals related to the process of scientific method and discovery are already sufficient. A most important deficiency is that we can nearly never find more extensive prototypical data analyses that show the students how they might answer to the questions in a deeper way. We may be biased in this respect because we found that the data the projects provide are so interesting and rich and many interesting aspects of data analysis and of the subject matter can be found out by means of data exploration. The current projects do not yet exploit this potential. To show what we mean we provide our own example of data analysis for each project, often trying to explore a question that the project posed itself. We intend to use elementary methods of data analysis but apply them in a flexibel and interactive manner in the spirit of exploratory data analysis (John Tukey)..
Our Summary is structured according to 1. The system; 2. Learning goals; 3. Available data; 4. Supports for data analysis; 5. Our own example of data analysis; 6. Summary from the perspective of data analysis.
1. We can conceive of all data sharing projects as refering to a system that is represented by a number of interacting variables and some context variables such as the water quality of a lake measured by several variables and context variables such as weather conditions. Or, the a road system that "produces" killed animals whose number and type depends on road type, weather conditions, time of the year and time of the day, for instance. 2. Learning goals refer to what students should learn about the respective system. These goals may be content goals, for instance some knowledge about water quality and basic chemical and physical processes, or typically rather process goals: students are encouraged and supported to behave like a scientist and participate in data collection and analysis as well as in the process of scientific research from developing and questions and hypotheses up to answering some of the questions by means of data collection and analysis. 3. Available data briefly sumarize what we discuss in the chapters on data archive and data. 4. supports for data analysis contain software tools, subject matter knowledge for data interpretation, data analytical and statistical knowledge and strategies, exemplary data analysis and expected answers and summarizes findings from the chapters D. and E. from a more general perspective. Our own example of data analysis (part of E.) is summarized in point 5. The last point 6. provides a short overall evaluation.
The fact that the summary differ in structure from the chapter structure is due to the need to enable comparison across projects (the aim of the summary) and to portrait each individual project and its specificities form the perspective of data analysis, which is the aim of the chapters B. to E.
3. The Five Technical Reports
2. Estuary Net
1. Journey NorthGlobe
3.
4. Roadkill
5. Water on the Web
4. Acknowledgements
We are very grateful to Cliff Konold and his co-workers at the University of Massachusetts who did a lot of work in improving the language and the understandability of our 5 technical reports. Anyway we are responsible for all errors and unclearnesses that do still exist. This research was supported by the National Science Foundation, Washington (NSF Grant No. REC-9725228).