In an effort to standardize the calculation of monthly child support awards, the federal government requires states to use preestablished formulas to determine the amount of awards. This study utilizes survey data from 20 states to determine the extent to which individuals who regularly calculate child support awards using these state formulas do so accurately. We find that individuals who regularly calculate child support awards often arrive at different support award amounts because of the complexity of the state guidelines, human error, differences in their experience and training, and lack of computerization. Adopting simpler formulas would reduce errors; this should be a priority even if child support officials use computers to calculate award amounts. Efforts to further the training and education of personnel who calculate awards would also help, and child support offices should revise their formulas to cover high-income cases. The Child Support Enforcement amendments of 1984 (P.L. 98-378) required states to develop specific numeric formulas for determining the amount of child support that should be paid to children by absent parents. This legislation was enacted because child support awards were typically too low (Beller & Graham, 1986, 1993; Oellerich, Garfinkel, & Robins, 1991; Williams, 1987), there was substantial variation in the amount owed by individuals in similar economic circumstances (White & Stone, 1976; Yee, 1979), and case-by-case award setting had resulted in administrative inefficiencies and had aggravated problems for interstate cases (Thompson & Paikin, 1985; Williams, 1987). The 1984 legislation, however, provided states with minimal direction in how to frame the content of the guidelines. Sensing this problem, Congress postponed the original implementation deadline from October 1985 to October 1987 (Brackney, 1988). The original 1987 state guidelines were based on three models that are known as the percentage of income, income shares, and Melson models. Including the District of Columbia, Puerto Rico, Guam, and the U.S. Virgin Islands, there are 54 unique variations of the three basic child support models today. When introduced in 1987, the state guidelines were advisory, meaning that individuals calculating child support awards could have used, but were not required to use, their state's guideline. The National Advisory Panel on Child Support Guidelines first recommended that state guidelines become presumptive rather than advisory in 1987. That is, the amount of a child support award calculated using each state's guidelines is presumed to be the correct amount of child support unless good cause for deviating from the guidelines can be established. The advisory panel urged the use of presumptive guidelines, citing the same reasons used to adopt guidelines in the first place: to further improve the adequacy of awards, to ensure greater equity for litigants, and to facilitate the adjudication of cases (Advisory Panel on Child Support Guidelines, 1987). The statewide application of child support guidelines became presumptive with the passage of the 1988 Family Support Act (P.L. 100-485). More recently, the U.S. Commission on Interstate Child Support (1992) recommended that Congress appoint a National Child Support Guidelines Commission no later than January 1995 to determine if the United States should move to a single, national child support guideline. This commission asserted that very different amounts of child support are required of obligors, depending on which of the 54 state guidelines is used and that this variation results in an inequitable interstate system. A federal child support guideline also is viewed as an essential component of a reformed child support system by Ayuda, Clinica Legal Latina, the Center for Law and Social Policy, the Children's Defense Fund, the National Women's Law Center, the United States Catholic Conference, and the Women's Legal Defense Fund (1994). …