The Internet has become a very valuable, if not indispensable, resource for both teachers and students of statistics. Advances in computers and software have made statistics more accessible to students in all disciplines. The Internet has carried this accessibility even further with resources such as tutorials and readily available data sets. For teachers, the Internet has made the delivery of statistics courses much more efficient.

This paper is divided into two main sections, resources for teachers and resources for students. This is not a clean cut dichotomy. Teachers as well as students will finds things of interest in both sections. However, the material in the resources for teachers section was selected to enhance the teaching of statistics, while the material in the resources for students section is more appropriate for those who are learning statistics.

The resources identified in this paper are focused on teaching statistics at the postsecondary level. However, some of the sources cited here could be used as resources for high school probability and statistics courses.

The Internet contains a very rich set of resources for teaching statistics. General resources such as course materials, problem sets, data sets, etc., can be found on the Internet. There are examples also of syllabus templates and actual interactive syllabi. Statistics teachers can learn how to make assignments, receive completed assignments, and do testing by use of the Internet. A statistics course can now be delivered completely over the Internet. Specialists such as educational statisticians can find resources specific to their area of interest on the Internet. The location of some of these teaching resources are identified and discussed below.

The Chance Database [19] is a good source of teaching aids and articles that would be useful for teaching an introductory statistics course. The Chance Database is maintained by The Chance Project, Mathematics Department, Dartmouth College. The materials in this data base were designed to help teach a "standard introductory probability or statistics course." Documents in Adobe Acrobat format are available for downloading from this location. This site also has links to other internet sources for teaching statistics.

A good location for statistics course resources is The Probability Web [16]. The Probability Web has Links to several probability related topics such as random numbers and Monte Carlo methods. There are also links to online journals and links to abstracts on journal articles related to probability. The Probability Web provides some very good support resources for statistics teachers.

Another good resource location for teaching statistics is the The Data and Story Library (DASL) [5]. This is "an online library of data files and stories that illustrate the use of basic statistics methods." A large number of stories and the corresponding data that help illustrate statistics concepts can be found here. The stories are listed by topic from archeology to zoology. The stories are also listed by method such as ANOVA, regression, correlation, etc. DASL is part of an "effort to enhance the teaching of statistics using computers." Teachers can submit stories to be included in the library.

Yet another important resource for statistics teachers is the StatLib Index [13]. StatLib is a service provided by the Carnegie Mellon University Statistics Department for the distribution of statistical software, information, and data sets. StatLib has a large list of statistics resources by topic. The topics include data sets, graphical models, multivariate statistics, and applied statistics. Here, there are also many links to other statistics teaching resources.

The World Lecture Hall [24], maintained by the University of Texas at Austin, is a worldwide listing of faculty who use the Internet to deliver class materials. The site is organized by subject and the statistics [25] page has links to several universities with online statistics materials. This is a good source for "course syllabi, assignments, lecture notes, exams, class calendars, multimedia textbooks, etc." The World Lecture Hall also has links to colleges and universities that have online courses in statistics.

For many teachers, an interactive syllabus is their first use of the Internet as a teaching tool. The amount of interactivity in a syllabus varies from just an interactive email address to a full-blown interactive syllabus with multiple links, a front page, and several other pages such as assignment details, data sources, samples of old tests, etc.

A teacher may want to look at samples of other syllabi before constructing his or her first interactive syllabus. A good approach is to use a syllabus template to begin building the online syllabus. A good template [2] to use for the first syllabus is found at Chaminade University, Honolulu, Hawaii. The HTML source code for this template has comments that explain how to modify the template for a particular course, teacher, and institution.

A good example of an interactive syllabus for a statistics course comes from the University of Tennessee Statistics Department. This syllabus [30] is for a graduate Survey of Statistical Methods course. This is an extensive syllabus. The syllabus has a detailed course outline and a diary of what was covered in each of the prior class periods. A student who misses class can find on the Internet a short abstract of the class lecture. The syllabus also has a list of all the students in the class and their major field of study. A discussion group forum is included so that students can use the Internet to chat with others in the class.

After developing an interactive syllabus, a teacher may want to use the Internet for assignments and tests. Tests and quizzes can be transmitted to and received from students by way of the Internet. Assignments can be completed in a likewise manner. Thus, the Internet can be used to increase the efficiency of the assignment/testing process. As an example, a syllabus [1] found at the University of Northern Iowa uses the Internet to provide answers to homework assignments. An interesting feature of this syllabus is a dual screen (HTML frames) which is used to show sample tests in one frame and the answers in another frame.

Many teachers also make copies of old tests available to students. Putting these old tests on the Internet is a good way to do this. A good example of an interactive syllabus with old tests online is found at the University of Tennessee Statistics Department site. The syllabus [28] here is for a graduate course in Applied Linear Models. The syllabus has links to an archive of old tests previously used in the course.

The next level in using the Internet as a teaching tool is to put the entire course online. In addition to the wide choice of materials available on the Internet for developing and teaching statistics courses, some statistics courses are now being taught entirely in an online format. For example, one such course is taught by David Olwell [15] at the University of Minnesota. The course is Statistics 1001, Introduction to Ideas of Statistics. This is a distance learning course and the online format makes sense. An interesting feature here is a recorded message by the instructor welcoming students to the course. The course syllabus also contains a list of Internet sites that are part of the course syllabus.

Also, statistics textbooks in HTML format are beginning to appear on the Internet. One such book is Statistics: The Study of Stability in Variation by Jan de Leeu [6] of the University of California at Los Angeles. This is a beginning to intermediate level statistics textbook with four parts and seventeen chapters. The the table of contents is hyperlinked to the different parts of the book. The book also contains a hyperlinked glossary with extensive examples and definitions of statistics terminology.

The Internet also provides a rich resource for statistics students. The student will find many opportunities for learning enhancement on the Internet. The student will be able to better grasp the concepts of statistics and complete his or her assignments more quickly by using these resources. The student will find an almost unlimited source of data sets and software tutorials on the Internet. For students studying specific statistical applications, such as educational statistics, the Internet provides resources that that are specific to the applications being studied. The locations of some of these student resources are identified and discussed below.

At excellent source for learning statistical concepts can be found at the Statistics Instructional Computing Laboratory (SICL) [23] at North Carolina State University. This site has demonstrations for a large number of statistical tests. For example, the mean, standard deviation, and sample size can be entered into a dialog box and the demonstration returns the confidence interval. Other tests such as T-tests and population proportions are demonstrated. Also, distributions such as binomial, normal, and Poisson are demonstrated.

A site that has extensive links (over 475) to statistics resources can be found at York University [8]. Here there are links to resources for solving almost any conceivable problem a statistics student would encounter.

In years past, one of the most frustrating things for students (and teachers too!) was the search for appropriate data sets that were needed to complete class assignments and research papers. Today, the Internet has solved this problem. The "classic" data sets that appear in statistics textbooks can now be found on the Internet. Also, there are hundreds of other data sets available for students to choose from. A particular data set can usually be found to fit a particular topic in statistics or to fit a student's interest. The sites listed below are some of the best sources for statistical data that are currently available on the Internet.

• Data Sets for Data Mining [27]
• Examples and Data Sets [7]
• Statistical Sources [18]
• Inventory of Humanities Data Sets Menu [11]
• Time Series Data Library [12]
• SICL Data Catalog Page [14]