Distance is an important and basic concept in geography. Many theories, methods, and applications involve distance explicitly or implicitly. While measuring the distance between two locations is a straightforward task, many geographical processes involve areal units, where the distance measurement can be complicated. This research investigates distance measurement between a location (point) and an area (polygon). We find that traditional polygon-to-point distance measurements, which involve abstracting a polygon into a central or representative point, could be problematic and may lead to biased estimates in regression analysis. To solve this issue, we propose a new polygon-to-point distance metric along with two algorithms to compute the new distance metric. Simulation analysis shows the effectiveness of the new distance metric in providing unbiased estimates in linear regression.