Previous studies of sexual size dimorphism (SSD) use a variety of size dimorphism indices (SDI's) to quantify SSD. We propose that a useful SDI should meet four criteria as follows; 1) it should be properly scaled, 2) it should have high intuitive value, 3) it should produce values with one sign, (positive) when sex A is larger than sex B, and the opposite sign when sex B is larger, and 4) it should produce values that are symmetric around a central value, preferably zero. Many previously published SDI's do not meet any of these criteria, and none meet more than three. We present an alternative SDI based on the mean size of the larger sex divided by the mean size of the smaller sex with the result arbitrarily defined as positive (minus one) when females are larger and negative (plus one) in the converse case. Careful selection of a primary size variable is crucial to meaningful interpretation of sexual size differences.