Variable-rate nitrogen (VRN) applications to corn (Zea mays L.) are attractive for possible economic and environmental benefits, but there are many factors affecting yield responses (YR) to nitrogen (N) fertilizer within fields. Soils supply a large percentage of N to plants, therefore, appropriate adjustments for the variable supply of N from the soils could help guide VRN applications.  The objective of this study was to characterize and classify spatial variability in corn YR to N fertilizer by using soil and terrain attributes. Seven 25-ha fields with no-till practices were planted to corn after soybean and received two N rates applied in the near-optimal range (112 and 140 kg N ha^-1) in alternate strips within each field during three years. Urea-ammonium nitrate solution was injected into the soil at V3-V6 corn growth stage. The fertilizer strips were harvested with a combine equipped with a yield monitor and GPS. Yield responses to N were calculated as differences between yields at the highest and lowest N rates in a grid pattern with cells ranging from 20 to 40 m. Because exploratory analysis revealed very strong spatial dependence in YR in all the fields, apparent soil electrical conductivity, differential elevation, slope, and topographic wetness index were used as covariates in the auto-logistic models for analyzing distributions of YR expressed as a binary variable (profitable and not profitable) within the fields.  The analyses indicated that no single covariate was significant in all fields. Across all fields, yield responses tended to be higher in areas of higher elevation and slope. Additional studies are needed to determine whether other site specific factors could be used to classify variability in YR within fields and delineate zones for VRN applications.