InformationWeek
Defining The Business Value Of Technology
|
November 20, 2000 |
 |
The Unexpected Outcome
By Rusty Weston
How many of us knew the election would be that close? As a journalist and researcher, I viewed Election Night with equal measures of shock and unease. You could feel the political heat ratcheting up on the pollsters--the ones in the network news pool and the ones reporting to the Bush and Gore teams. The media, the politicians, and the public will demand substantial process reforms in the wake of this election. At long last, the elitist and outmoded Electoral College will be reconsidered. A chorus of pundits will suggest that both the voting and the tabulations should be available on the Internet. I believe the online world is ready for digital ballots, although many politicians, including California Gov. Gray Davis, do not. But the best reform idea I've heard yet is a simple, low-tech one: leaving the polling places open all weekend, rather than on a Tuesday.
Undoubtedly, the networks will think twice in the future before trying to predict outcomes in races better described as "too close to call." But there's an unexpected legacy of this election--one that concerns its possible impact on management practices. Let's describe it as a "reassessment" of the science of polling and forecasting outcomes. Realize that the business of judging probabilities is one that affects insurance, financial services, construction, sports, entertainment, and agriculture, among many other fields. This election has merely underscored what economists, gamblers, meteorologists, and actuaries long ago realized: There are limits to our understanding of the science of probabilities.
Thanks to this election, pollsters are going to have worse reputations than journalists, politicians, and weathermen. Madison Avenue is in the vanguard of social criticism with a pizza ad on TV that congratulates both George W. Bush and Al Gore for winning Florida and thus the election. It's a good laugh, but who is the joke on anyway?
Few people would say the public has a good handle on the science of sampling--exit polling or otherwise. But I'll bet that you're expected to make critical decisions based upon scant knowledge more than a few times a year, maybe even a couple times a week. A few examples that spring to mind: How many IT managers who oversee help desks can accurately infer from a few complaints whether a larger problem is at hand? How many people in your organization know how to reliably forecast Web traffic or product demand based on a new marketing campaign? You may be charged with sampling your users or customers to test the usability of your supply chain or E-commerce site. Is a focus group of eight people sufficient to influence your product-development plans?
You may find such forecasting efforts intuitive and easy. But what happens when you get it wrong? Will senior managers believe you when you return to them seeking additional resources based upon business predictions that may be difficult to swallow? (A poor track record of forecasting is generally not the way to build a career.) You can always hire a highly paid consultant to help you read the tea leaves. Or you can read a book about statistics--or you can hire a highly paid consultant. Oh, wait, I already mentioned that option.
It turns out that proper sampling is like working up a spreadsheet. You've heard the phrase "garbage in, garbage out." We've probably all seen (or written) business plans that include a range of revenue targets based upon forecasts that lack empirical evidence. How many businesspeople (or MBA students, for that matter) take the time to properly sample a market before seeking venture-capital funding? In what must be a surprise to many novice managers, it turns out that understanding a market demographic isn't the same thing as sampling potential customers. Tastes do change, even if the demographics remain the same.
Among its myriad legacies, Watergate spurred many students in the '70s to take up journalism as a career. In the wake of the presidential election, I wonder if the negative publicity will have a chilling effect on students who might otherwise have been drawn to research. I hope not. Because it's unrealistic to expect most business or IT people to master statistics and probabilities. Still, some people believe research is just an arcane discipline or an unnecessary cost. And I've got a centrally planned economy in Europe to sell to these geniuses.
Rusty Weston, editor of InformationWeek Research and InformationWeek.com, is a former card-carrying member of the American Statistical Association.
Authorities Archive
Send Us Your Feedback
Top Of The Page
 | |||
| Lou Bertin: The Observer Lou offers a view of the good, the bad, and the bizarre developments in the technology business | Charles Pelton: Eye On IT Charles explores IT management issues and strategies that business and technology managers face. | Jason Levitt: Internet Zone Jason focuses on the strange, egregious, and the standard technologies of the intranet/Internet. | Stuart Johnston: Redmond Watch As our eyes and ears in Redmond, Stuart gives his perspective on the latest events at Microsoft. |
ACCO Brands Corp seeking Director of New Product Development in Lincolnshire, IL
Transportation Security Administration seeking Chief Information Officer in Arlington, VA
Hebrew SeniorLife seeking Business Systems Analyst in Boston, MA
Trilogy Leasing seeking General Manager in Cranbury, NJ
UVIMCO seeking Senior Information Technology Leader in Charlottesville, VA
For more great jobs, career-related news, features and services, please visit our Career Center.
Today's Top News
- Mobile E-Mail For Cost-Conscious Enterprises
- IBM, Second Life Demo Virtual World Interoperability
- Chambers Sees Cisco CEO Role Changing In Five Years
- Broadcom, Qualcomm Face Off In Chip Hearing
- Verizon Drops 1938 Media After Charges Of Racism
- Microsoft Patch Tuesday Brings Four Bulletins For Nine Flaws
