Creativity is understood intuitively, but it is not easily defined and therefore difficult to measure. This makes it challenging to evaluate the ability of a digital tool to support the creative process. When evaluating creativity support tools (CSTs), it is critical to look beyond traditional time, error, and other productivity measurements that are commonly used in Human-Computer Interaction (HCI) because these measures do not capture all the relevant dimensions of creativity support. Unfortunately, there are no clear measures of success to quantify in regards to creativity support tools, and this lack of 'convenient' metrics is a real challenge to their evaluation.In this dissertation, I introduce two computational methodologies for evaluating creativity support tools, including: (1) the Creativity Support Index (CSI), which is a psychometrically developed and validated survey, designed for evaluating the ability of a tool to support the creative process of users, and (2) a novel sensor data approach to measuring 'in-the-moment-creativity' (ITMC), to detect moments when users experience high creativity using electroencephalography (EEG), activity metrics (e.g., keyboard/mouse logger and accelerometer data), and machine learning.