Double-cropping of soybeans and wheat is often promoted by extension personnel. This paper seeks to explain how the decision to adopt double-cropping is made, using a Tobit regression model. Tobit makes use of more of the information in the data set than do logit or probit and explains not only the decision to double-crop but also the rate of adoption. The paper considers factors such as profit and risk perceptions and risk which have not been included in the past models used to explain adoption of technology. The results show that risk perception is important. Contrary to the findings of some other adoption studies, this decision is not influenced by human capital factors. The farmers who double-crop are more highly leveraged and appear to do so both to achieve higher income and as part of a risk diversification strategy. This is consistent with the importance of the location factor, measured as the average number of growing degree days at the farm's location. Growing degree days is a proxy for the actual distribution of returns from double-cropping and is the main factor explaining this decision. Extensive adoption of double-cropping in cooler regions of the Midwest must await technological advances that can increase the profitability of double-cropping by reducing the growing season for wheat and/or beans.