SPRIG sponsored two research sessions at the INFORMS Atlanta Conference in October 2003. An 80-seat room was filled beyond capacity! Clearly Spreadsheet Productivity is a growing and vital area. Unlike many sessions at conferences, the audience was diversified across government, industry, and academic segments.
The following presentations were made at the SPRIG-sponsored sessions. Some of these presentations are now available on the SPRIG website. Alternately, contact the authors for more information.
Besides the SPRIG-Sponsored sessions, there were several other Spreadsheet-oriented sessions. Go to http://www.informs.org/conf/Atlanta2003/program.htm to search the full Atlanta program.
Sixty-seven people signed up as new members of SPRIG at the INFORMS Atlanta Conference. Welcome! Consistent with our mission, members span government, industry, and educational sectors.
For specific information about SPRIG-sponsored sessions, or to suggest a session and/or paper topic, contact the SPRIG VP-Meetings, Jenny Wagner, at janet.wagner@umb.edu.
$1.2 Billion Spreadsheet Error at Fannie Mae (New York Times, 10/30/03). Note: The full text is not available free from NY Times. If your organization has an account with Lexis-Nexis, it is available there. A pertinent excerpt from the article:
"In adopting a new and complex accounting standard in a short period of time, Fannie Mae had to put in place a system and process to capture all open commitments and mark them to market," the company said in a statement. "To implement this standard, Fannie Mae utilized information from its internal, automated systems in conjunction with spreadsheets that made additional calculations necessary under the new rule." A spokeswoman said that one of those spreadsheets contained an error (emphasis added)."
The Spreadsheet Page, J-Walk and Associates. This site has tons of useful tips and tools for the spreadsheet developer.
Slicing Spreadsheets: An Integrated Methodology for Spreadsheet Testing and Debugging. This paper by James Reichwein, Gregg Rothermel, and Margaret Burnett was presented at the 2nd Conference on Domain-Specific Languaes, October 3-5, 1999, Austin, Texas. From the abstract: "To accommodate the users of spreadsheet languages, we provide an interface to our methodology that does not require an understanding of testing and debugging theory, and that takes advantage of the immediate visual feedback that is characteristic of the spreadsheet paradigm."
From Raymond Panko's paper, What We Know About Spreadsheet Errors, available at http://panko.cba.hawaii.edu/ssr/Mypapers/whatknow.htm. Ed: Not only do almost all studies suggest high error rates in spreadsheets, but most spreadsheet developers are overfident about their spreadsheet's correctness. This quote suggests the problem is bigger than spreadsheets.
"While such massive levels of overconfidence seem unreasonable, overconfidence is perhaps the most well-established phenomenon in the behavioral sciences. When asked about many things, such as driving, a large majority of all people rate themselves as above average (Brown, 1990; Gilovich, 1991; Koriat, Lichtenstein, & Fischoff, 1980). A long stream of experiments have found that people are overconfident in a wide variety of tasks ranging from Trivial Pursuit type questions to the perception of acoustical signals (Adams & A., 1961; Lichtenstein, Fischoff, & Philips, 1982; Plous, 1993; Wagenaar & Keren, 1986). Overconfidence has also been seen in decision support systems use (Kasper, 1996) and database queries (Greenblatt & Waxman, 1978). Even experts in a wide variety of fields are often appallingly overconfident within their domains of expertise (Johnson, 1988; Shanteau & Phelps, 1977). Overall, in both contrived and realistic tasks, there seems to be a pervasive human tendency toward overconfidence."
SPRIG is administratively housed in the INFORMS (Institute for Operations Research and the Management Sciences) professional society. INFORMS (http://www.informs.org) provides valuable services, including web hosting, email list management, and support for SPRIG sessions at their annual conferences.
Everyone who signed up for the SPRIG email list will automatically be a member of the SPRIG section of INFORMS (but not of INFORMS itself, if not already a member). For the year 2004 this will be free of charge. In future years, SPRIG will charge nominal dues that will help defray expenses of SPRIG activities. If for some reason this does not work for you, but still want to receive the email newsletter, contact the SPRIG secretary (link provided below) and you will continue to receive it.
Please contact any one of us if you have any questions, comments, or suggestions!
This newsletter is published several times per year. Submissions are strongly encouraged. Please send ideas, suggestions, and links to the newsletter editor (roger.grinde@unh.edu). Thanks
To be added to or removed from this SPRIG email list, please contact the SPRIG Secretary at jeff.keisler@umb.edu. Past newsletters are archived at http://sprig.section.informs.org/.