SBIR Phase II: Large-scale Creative Thinking Assessment for the Workforce

Sparkting, Berkeley CA

Investigators

Abstract

This SBIR phase II project is focused on quantifying and developing one's creative thinking abilities. In more details, this project will develop modules that measures one's domain-base and subject-base creativity and validate the modules using statistical validation techniques. Creative talent will remain the great differentiator in the coming world, and the organizations that thrive there will do so because they have the right potentials. Using the semantics-based psychometric approach, this project designs exercises and models that assess one's creativity. This project will help employers in acquiring creative employees and also train and educate current employees in ways by which they can become more creative. The market is vast, encompassing potentially all employers who hire and manage workers with symbol-analytic skills. According to the 2010 U.S Census, there are over 143M people employed in the workforce. This project will expand to various sectors of the workforce using strategic partnerships with distribution channels in workforce education companies, and talent/human capital management companies. Currently, creative thinking strategies and assessments are taught by experts and consultants in the form of workshops. Such semantic creativity assessment is a manual, time-consuming and expensive process. This project creates a set of open-ended exercises and evaluates each exercise based on the four creative-thinking dimensions: 1) Originality: Original thinking capacity and ability to generate novel and out-of-the-box solutions; 2) Fluency: Ideation capacity and ability to push past the first set of known responses; 3)Flexibility: Divergent-thinking capacity and ability to think non-linearly; and 4) Elaboration: Detailed-oriented and ability to provide intrinsic details about each possible solution and response. This project automates such testing by using advances in semantic-based psychometric modelling, natural language processing (NLP), semantic networks from computational linguistics and computational power for statistical mining of large corpora. Due to ease of scalability, this solution is not limited to any one country or region. The solution can be deployed world-wide. Currently, this project has gathered data from 161 different countries around the world. This project is the first step towards automation of open-ended exercises in various fields and contexts such as situational judgement assessments, emotional intelligence and motivational assessments.