This article evaluates the suitability of tobit analysis of food consumption patterns by comparing results generated by a tobit model and a modification with a two‐step decision process. Two samples of adult women were selected from the 1977–78 Nationwide Food Consumption Survey and the 1985 Continuing Survey of Food Intakes by Individuals. Results show strong statistical evidence for questioning tobit analysis in examining food consumption decisions. Determinants of the decision to consume food within a particular food group often differ from the determinants of how much to consume, particularly for more highly defined food groupings.