Conclusion: There are many advantages to treating variables as continuous numeric variables if
the data allow this, and this should be employed more often in experimental biology. Failure to use
the optimal analysis runs the risk of missing significant effects or relationships.