The results in table 4 include the values of the logit coefficients, the t-statistics, and associated p-values. Figure 1 portrays the probability response to changes in the explanatory variables.Only corn results are reported—the soybean results are very similar with the few notable differences indicated below. Resultswere more statistically significant with the use of
the (a) debt-to-asset ratio instead of debt use; (b) the perceived probability of receiving an APH yield insurance payment instead of the county-level insurance premium; and (c) the use of the average overall risk management
score instead of the use of specific risk management options (e.g., hedging/options and forward contracting scores) alone. In each case, the correlations between the alternative representations of the independent variableswere relatively high, indicating that use of only one of the two measures is preferred. The measures yielding stronger levels of significance are retained throughout the analysis.