Newsletter for the Section on Statistical Education
Volume 3, Number 2 (Summer 1997)
Editor's note:We highly recommend your viewing of the html version of this article at the address: http://redrhino.mas.vcu.edu/rein/StatEd/ since it contains direct links to the web pages mentioned in this article.
It appears that we are always just around the corner from truly using technology and the internet to aid us in the teaching of introductory statistics. Well, I believe we may very well have turned the corner, perhaps without even knowing it.
With tools such as the internet, hypertext and web browsers, virtually anyone can self publish their opinions, ideas and computing tools. For example, please see "Statistics Every Writer Should Know'' by Robert Niles (reference given below) . Because of this, there is a great wealth to be found out there on the Web ... and much that is truly worthless. For novices to be able to properly sink their teeth into the Web and its resources, a simple pamphlet-like guide to resources of interest appears to be necessary. This is my attempt to provide such a guide for statistics education resources.
I've attempted to categorize these resources by content but should mention that there is a fair amount of overlap across categories: Statistics Lists of Lists; Statistics Education Research; Case Studies, Datasets, and Stories; Computers as a Calculation/Demonstration Tool; Computers as a Communication Tool; and Other, more general resources. URLs (essentially addresses for each webpage) are given in the reference section where items are listed hierarchically by webserver.
Although there is much that is valuable on the World Wide Web, the key to finding what will be valuable to you is to find a good solid reference. Within the discipline of statistics, there appear to be two such starting points that are a bit more thorough than the others, StatLib and the Statistics entry of the WWW Virtual Library.
StatLib is the creation of Mike Meyer and resides at Carnegie Mellon University. Essentially, it provides links to many of the major statistical resources on the World Wide Web. StatLib also happens to be the place where The Data and Story Library (more on this later) is housed.
The Statistics entry of the WWW Virtual Library is a similar list of web resources for statisticians and those interested in statistics. Both contain links to publicly available datasets.
Much information about the teaching of statistics is available on the Web, particularly through the Journal of Statistics Education. The Journal of Statistics Education describes itself as "an electronic journal on post-secondary teaching of statistics.'' There is much useful information here including an article by Joan Garfield entitled "Teaching Statistics Using Small-Group Cooperative Learning''. (Another excellent article by Garfield that is available on the Web is "How Students Learn Statistics''; see references.)
The Chance course -- where the aim "is to make students more informed, and critical, readers of current news that uses probability and statistics'' --provides through the Chance webpage a list of Teaching Aids (including Data Sets and Programs that a browser can run) as well as Chance News, "a biweekly newsletter providing abstracts of current news items of statistical interest.''
One of the pages in the StatLib hierarchy is The Data and Story Library which indexes publicly available datasets by topic, statistical method, and data subject matter. The DASL materials not only allow one to quickly search for datasets by topic or method, but by using the powersearch facility, one can look for datasets and stories that relate to one's location (for example, I just found out that Richmond, VA has about 40 inches of rain per year and a mean July temperature of 78).
The Case Studies page at UCLA has twenty-some real life uses of statistics. For each there is a description of a real life problem, some data, and a few questions appropriate for beginning students. There are also answers provided. UCLA also has a large collection of links to other data sources.
Although few instructors have at this time the resources necessary to access the Web during class time, the following links may be useful if you are creating a webpage for your class and want the students to be able to "try it out themselves''. The Chance webpage has a list of Programs That A Browser Can Run, Duke keeps a list of Java Applets, and UCLA has a list of Statistical Calculators. UCLA also has an excellent collection of Xlisp-Stat Demos (which require Xlisp-Stat to run, but that is freely available). There is also an Xlisp-Stat Archive.
Computers are no longer limited to performing calculations and displaying graphs in their role as supporting statistics education. They are also wonderful for communication. The Web is proof of this!
5.1. Virtual Benchmark Instruction and HyperNews. Andrew Schaffner, David Madigan and others at the University of Washington have done some work on Virtual Benchmark Instruction (VBI). Essentially, a group of students will jointly tackle problems in an online discussion group. The problems are designed to (re-)emphasize key concepts in the course. Through the interaction the students will clarify much of their own misunderstandings. Details can be found on the Statistics Education Research web page. HyperNews is the software used at the University of Washington for VBI but can also be used for any sort of online class discussions. It should run on most webservers.
5.2. Usenet and E-mail Lists. Of the thousands and thousands of usenet newsgroups, two stand out as potentially valuable for teachers of statistics: sci.stat.edu and sci.stat.math. Sci.stat.math may be of interest to those teaching more advanced courses as the content is a bit theoretical in nature and certainly less aimed at issues of statistics education than is sci.stat.edu.
Sci.stat.edu covers the teaching and learning of statistics and contains exactly the same discussions as does Listserv (e-mail list) EdStat-L. Literally, the usenet group is linked to the Listserv group so that any articles posted to the usenet site are automatically e-mailed to the Listserv members and any e- mails sent to EdStat-L are posted to sci.stat.edu. If you don't have access to usenet or prefer e-mail over usenet, you may want to subscribe to EdStat-L.
To subscribe to EdStat-L, simply send an e-mail to firstname.lastname@example.org where the body of you e-mail should read: subscribe edstat-l YOUR NAME (where, of course, you would replace "YOUR NAME" with your name).
There are literally hundreds of lists of lists and links to links on the Web. The two that I've personally found the most helpful in general are Alta Vista and Yahoo!.
Yahoo! looks like an index. From the home page, one can select "Society and Culture'', "Government'', "Reference'' or any of several other categories. Under each of these categories are several others which somewhat sub-divide the main category. For example, under "Society and Culture'' are "Museums and Exhibits'', "Race Relations'' and others. On each page, Yahoo! lists appropriate sub-categories and web pages. Yahoo! also has a search facility that typically will give some excellent starting points within the category hierarchy for any particular topic.
Alta Vista is substantially different. While Yahoo! is a list of lists and fairly easy to use, Alta Vista could be thought of as a organizationally challenged librarian. You can approach Alta Vista with a search term (or terms) and the result should be (in theory) a listing of links to all the pages (sometimes thousands) that include this term. It is often useful to combine search terms to narrow the results a bit. The results are ordered by a score which relates to how well each of the documents contains the search terms. This ordering, however, is not much like the ordering that a real human would give the list as it doesn't group the results into recognizable categories. Alta Vista does allow you to fine-tune your search by adding new words or by restricting the results to exclude particular words (or even domains).
For further information contact Steven Rein at Virginia
Department of Mathematical Sciences
Richmond, VA 23284-2014
(804) 828-1301 x136