My research is a blend of thinking and doing, of the empirical and the constructive.
|On How to Manage by Designing|
|With our 2002 workshop on Managing as Designing we initiated a dialogue that brought a broad diversity of designers together with management scholars to explore how practices from design could inform management. This workshop resulted in a documentary film, a book, a series of ongoing international gatherings, and major changes in the curricula of Weatherhead and other management schools.|
Richard J. Boland, Jr. and Fred Collopy, Design Matters for Management, from Managing as Designing, Stanford University Press, 2002, 3-18.
Fred Collopy, I Think With My Hands: On Balancing the Intuitive and Analytic in Designing, in R. J. Boland, Jr. and F. Collopy [eds.], Managing as Designing, Stanford University Press, 2004, 164-168.
Richard J. Boland, Jr. and Fred Collopy, Toward a Design Vocabulary for Management, in R. J. Boland, Jr. and F. Collopy [eds.], Managing as Designing, Stanford University Press, 2004, 265-276.
Richard J. Boland, Jr., Fred Collopy, Kalle Lyytinen and Youngjin Yoo, Managing as Designing: Lessons for Organization Leaders from the Practice of Frank O. Ghery, Design Issues, 24 (2008), 10-25.
On Representing Financial Data
|Using the cycle model, an abstract representation of the operations, financing, and capabilities of a firm, we developed and tested a program for viewing basic accounting and financial information visually and dynamically. In experiments we have found that subjects using it are able to identify which firms are likely to go bankrupt more accurately than those using spreadsheets or conventional graphs.|
Demonstration software and a paper describing the experiments are available.
Lin Zhao, Julia Grant, Fred Collopy and Richard J. Boland, Jr., Dynamic Representation of Financial Ratio Data: Designs and Empirical Results.
Richard J. Boland, Jr., Fred Collopy, Julia Grant, and Lin Zhao, Virtual Prototyping of Financial Flows as a Form of Management Control, in Brandon and Kocaturk [eds.], Virtual Futures for Design, Construction and Procurement, Wiley Blackwell Publishing, 2008.
On Visual Instrument Design
|I designed and programmed Imager, a performance instrument that provides real-time control of color, form, and motion of abstract images so that I can play with abstract images as musicians play with sounds.|
Fred Collopy, Color, Form, and Motion: Dimensions of a Musical Art of Light, Leonardo, Vol. 33, No. 5, 2000, 355-360.
Fred Collopy and Robert M. Fuhrer, A Visual Programming Language for Expressing Visual Rhythms, Journal of Visual Programming Languages, 12, 2001, 283-297.
Fred Collopy, Robert M. Fuhrer, and David Jameson, Visual Music in a Visual Programming Language, IEEE Symposium on Visual Languages, (1999), 111-118.
Fred Collopy, Improvisational Lumia: Playing Along with Musicians, Leonardo, Vol. 34, No. 4, 2001, 353.
On the Use of Information and Information Systems
In a series of lab studies and retrospective field research, we found that competitor-oriented objectives and some uses of competitor-oriented information can be detrimental to profits.
J. Scott Armstrong and Fred Collopy, Competitor Orientation: Effects of Objectives and Information on Managerial Decisions and Profitability, Journal of Marketing Research, 33 (1996), 188-199.I identified systematic biases (regression to mean) in managers' self-assessments of the amount of time they spent using computers. At least then, heavy users thought they used them less than they actually did; while light users perceived their use as greater than it was.
Fred Collopy, Bias in Retrospective Self-Reports of Time Use: An Empirical Study of Computer Users, Management Science, 42 (1996), 758-767.We defined and tested nine principles to support the design of user-tailorable technologies.
|On Business, Demographic and Economic Forecasting|
We developed Rule-Based Forecasting, an expert systems approach to improve the selection and combination of extrapolation forecasts and in integrate judgment with statistical methods.
We found that machine learning techniques could improve on the estimates of experts for coefficients used in expert systems for forecasting and that neural networks could be effective under very particular conditions.
We reviewed the approaches proposed for integrating statistical and judgmental forecasts and the empirical support for each.
We found that of 48 studies that examined the use artificial neural networks to produce forecasts, only 22 were effectively validated and implemented. Of those, 18 supported the potential of neural nets for forecasting and prediction.
We proposed and tested the use of causal forces for the selection and weighing of extrapolation methods and then for decomposing time series.
We found that the most widely used measures for assessing accuracy in forecasting studies are unreliable and unstable and proposed and evaluated the Relative Absolute Error (RAE) as an alternative.
We found that prediction intervals are often asymmetric and proposed a method for modifying them.
We determined that the use of diffusion models to forecast information systems spending has produced errors that are greatly in excess of those resulting from the application of simple extrapolation methods. We also developed some important principles for conducting comparisons of forecasting methods.