Evaluation of data transformations used with the square root and Schoolfield models for predicting bacterial growth rate

The square root and Schoolfield models are equations used to predict bacterial growth rate as a function of temperature. Several variations of these two models and the resulting statistical consequences were evaluated. The results showed that growth rate variance increased as a function of temperature. To compensate for the increased probable error, the natural logarithm transformation was more effective than the square root transformation. However, when natural logarithm transformations were employed for both models, the square root model was more accurate than the Schoolfield model.