Enantiomeric fractions (EFs) are used extensively in environmental pollutant research because of the insights on biochemical weathering available from quantifying enantiomeric composition. While this analysis is powerful, it can also be subject to significant error, depending on how chromatographic peaks are integrated. Two methods of integration, the common valley drop method (VDM) and the deconvolution method (DM) were compared using both instrumental and simulated chromatograms to assess their performance when integrating pairs of enantiomers. The effect of peak parameters such as true EF, peak resolution, signal-to-noise ratio, and asymmetry were also investigated. The VDM biased EFs by up to +6% to -4% (relative to the 0-1 EF scale) for symmetric peaks, and as low as -20% for asymmetric peaks. For both instrumental and simulated data, biases tended to increase with decreasing resolution and more extreme (nonracemic) EFs. In contrast, the DM produced biases that were less than 1% in most cases, including at very low resolutions. Estimates from previously published studies based on EF, such as biotransformation rate and source apportionment, could be dramatically affected by small errors in EF. Our results suggest that a deconvolution-based integration method is preferable for the handling of enantiomer compositions. Caution is also advised when comparing published studies on chiral environmental pollutants as most do not specify how chromatographic data is processed.