Generalized additive models (GAM) and spline regression models add smaller basis splines together to make a larger spline that fits as closely as possible to the empirical data. Hence, the coefficients in the resulting spline regression model represent the degree of scaling applied to each basis spline [18]. This makes it difficult to interpret the resulting regression coefficients in terms of the observed data since they are calculated for several smaller splines that are scaled between each knot making up the curved line that is being modeled. In other words, the calculated spline coefficients are in fact scaling coefficients that do not necessarily reflect the bacterial growth rates. This is also true for linear b-splines (splines of polynomial degree = 1), as these splines consist of multiple lines joined together as curves. Thus, each basis spline, linear or non-linear, may have many derivatives for each knot since many lines/curves make up a basis spline. Luckily, statistical programs are available that can calculate the slope of the joint line segments, as opposed to the scaling coefficients, so that time consuming multiple linear regressions will not have to be performed. In the present work the bacterial growth curves were compared using a standard linear spline regression model of which the slope of the growth estimates could be obtained. The slopes were subsequently compared using t-distributions, which allowed us to check whether the growth rates, i.e. the calculated slopes of the different growth curves, increased or decreased when compared between the different strains. Although the t-distribution estimated the confidence intervals for the data used here adequately, other distributions or bootstrap re-sampling can be used instead, should the data demand it. This procedure makes it both easy and fast to compare many bacterial growth curves with statistical certainty.