This report, prepared for the National Science Foundation, describes the process we used to develop a tool that predicts a project's outcomes from its proposal. We defined proposal characteristics and outcome extents for 36 completed educational software development awards. We used multiple regression analysis to specify and fit a predictive model, and cross-validation to assess the predictive ability. Finally we developed a software tool that allows program officers to apply this model to help evaluate new proposals.

• This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.

Creator

• White, Alexander Lee
• Messier, Kelly D.
• Caulkins, Robert John

Subject

• Computing
• Learning

Publisher

• Worcester Polytechnic Institute

Identifier

• E-project-121004-094416

Keyword

• Application software--Development

Advisor

• Demetry, James S.
• Petruccelli, Joseph D.

Year

• 2004

Center

• Washington, DC Project Center - IQP

Sponsor

• National Science Foundation

Date created

• 2004-12-10

Location

• Washington

Resource type

• Interactive Qualifying Project

Rights statement

• In Copyright