Intra-examiner reproducibility of the agar plate method for estmation of phospholipase activity in candida albicans

Abstract

Although measurement errors can impair statistical analysis, reliability analysis has been neglected in applied microbiology. This study assessed the intra-rater reproducibility of the Agar-based method for estimation of phospholipase activity (Pz). Pz readings were performed twice by two examiners (E1, E2), either directly on plates or in photos, and both black and white backgrounds were used. Pz values were taken from one or triplicate colonies from each sample (n=30). Intra-examiner reproducibility was estimated using Intraclass Correlation Coefficient (ICC). For both examiners, reading triplicate (ICCE1=0.91, ICCE2=0.86) was better than one colony (ICCE1=0.86, ICCE2=0.80). E1 had an excellent concordance when measurements were performed on photos using a white background (ICC=0.95) and a good concordance in the other conditions