Food selection by a planktivorous fish, the gizzard shad, Dorosoma cepedianum, was studied in Tuttle Creek Reservoir, Kansas. To determine if selection was occurring a fish stomach sample and a plankton sample were taken concurrently at 22 separate sampling times during the study. Percent composition of the various food items in paired stomach and plankton samples was compared by an electivity (E) index. Four least squares analyses of variance were performed on the E index values to determine which factors (i.e., species of plankton, date of collection, time‐day or night, size‐class of fish or size‐class of plankton) explained most of the variation in E and thus was responsible for selection by gizzard shad. Differences between the species of plankton in the analysis was found to be the most consistent factor contributing to selection. Differences in size, mobility, behavior, and so forth might be considered as causative agents responsible for positive or negative selection by gizzard shad. Of these species differences, organism size was shown to be of some importance. In addition to species (plankton), the other factors mentioned appear to contribute to selection, but to a lesser extent. If young gizzard shad are selective in their feeding, then the impact of the shad spawn should be observable in changes in zooplankton species composition and population size during the time when the carnivorous life history stages of this fish are present in the environment. To test this hypothesis the zooplankters were assigned to “selected for” and “selected against” groups. The “selected for” group declined to much lower density than did the “selected against” group. The decrease in zooplankton density, particularly the marked decrease in the “selected for” group, supports the hypothesis that selective predation by young shad has an impact on zooplankton populations.