Abstract
Traditionally, when performing a spectroscopic measurement, temperature and path length are normally set by modifying one factor at a time. This fact means that spectroscopists can be unaware of some possible interaction between these factors and it supposes that a lot of experimental effort is required before inexperienced researchers understand the extent of the variation these may introduce. In this study, experimental design methodology and the subsequent evaluation of results were proposed and applied to this problem in order to make the selection of conditions easier and ensure that it is performed correctly. The present paper is a case study which focused on the selection of path length and temperature for near infrared measurements of vinegar samples from different origins. The main objective was to ensure repeatability between replicates and to make sure that absorbance values in the spectra fell within the linear absorbance range of 0.5–2 au. This paper reports the detection of possible interactions between the factors which can lead to erroneous interpretation of results.