I am Ryan Watkins and as a professor at George Washington University, where I do research on and teach about needs, needs assessments, and how people collaborate in making decisions with increasing ‘intelligent’ technologies.
Making judgments about what actions to take (or what actions to recommend that others take) routinely requires complex considerations about the desired and undesired consequences of those actions. These considerations are most commonly derived from a combination of (a) the goals we wish to accomplish but have not yet achieved, along with (b) which actions are optimal for achieving those goals. Evaluators often refer to the processes associated with these decisions as Needs Assessment. And we define the “gaps” between desired and current results as ‘needs’, while the activities to close those gaps can be considered possible ‘satisfiers’.
Along with the rest of the world, needs assessment (as a field) is evolving in parallel with the data-driven technologies that support and guide more (and more) of our decisions. Within this context I would like to suggest that the construct of needs, and the processes for assessing needs, can and should be increasingly incorporated into the design and implementation of advanced technologies within the general label of “artificial intelligence” (AI). Since many of the advances in AI over the last twenty years have been gleaned from improvements in machine learning, this has increasingly focused the technology on predictive algorithms — which only informs one part of the judgments the people make (and increasingly machines help guide) on a daily basis. Prediction alone, however, is not enough. The concept of needs (which incorporates the concepts of necessity and sufficiency) is also extensively utilized in human decision making, along with the closely related concept of wants; though these are not yet part of the machine ‘intelligence’.
The more formalized integration of needs, I propose, can be valuable at multiple stages of AI design, development, and implementation. From initial human-centric decisions on what needs a proposed AI system might address, to complementing predictive algorithms when prioritizing options the AI system might recommend to human users, the fundamental construct of needs and the underlying measure of needs from the needs assessment literature can help improve the quality of AI aided decisions in the future.
Why should evaluators care? I believe that evaluators and our experiences with applying needs assessment from our unique perspective can (and should) be more than just consumers of AI technologies — we can be contributors too. Our foundational constructs and time-tested processes can help inform the development of the ‘intelligent’ tools that people are increasingly relying on more for support in their decision-making.
- Article on the value of necessity and sufficiency in explainable AI: Local Explanations via Necessity and Sufficiency: Unifying Theory and Practice
Article on needs in the development of Augmented Reality (AR):A Needs-Based Augmented Reality System
The American Evaluation Association is hosting Needs Assessment (NA) TIG Week with our colleagues in the Needs Assessment Topical Interest Group. The contributions all this week to aea365 come from our NA TIG members. Do you have questions, concerns, kudos, or content to extend this aea365 contribution? Please add them in the comments section for this post on the aea365 webpage so that we may enrich our community of practice. Would you like to submit an aea365 Tip? Please send a note of interest to firstname.lastname@example.org. aea365 is sponsored by the American Evaluation Association and provides a Tip-a-Day by and for evaluators. The views and opinions expressed on the AEA365 blog are solely those of the original authors and other contributors. These views and opinions do not necessarily represent those of the American Evaluation Association, and/or any/all contributors to this site.