Abstract
Leave a comment or question for the poster presenter below their abstract.
Especially in IT, it is difficult to escape the dreaded "hours estimate." Even in areas without a formal timecard or billing requirement, managers and directors often want to know how much time is being spent on different projects to inform planning and resource allocation. Yet how often does estimating project hours become a burdensome memory exercise and unnecessary headache, especially in a fast-paced IT environment where multitasking and remote work is the norm? Thankfully, statistics can come to our aid. In this poster, we outline an analytical approach that reduces hours estimates to a simple choice of "T-shirt size" with minimal sacrifice of accuracy. In our implementation within HITS Academic Integrations, we were able to estimate our actual project hours to within 16% with the appropriate tuning of our T-shirt size ranges based on simple statistical estimation methods.