This paper highlights construct validity issues that can occur when researchers utilize index variables (commonly referred to as ratios) in which the denominator can become negative. These indexes are commonly used in management research, and we demonstrate that indexes constructed from data which includes negative values for the denominator lose their ordinal properties and can be difficult, if not impossible, to interpret. Further, we show that common empirical practices to resolve these issues do not solve issues of ordinality and interpretability and may lead to additional validity issues. We recommend the use of alternate measures not affected by the issues described in this article to increase the validity of future organizational research.