Reliability of a two-dimensional footprint measurement approach

Abstract

Although footprint evidence can be taken from the scene of a crime, the science underpinning such measurement in forensic science has not been fully explored. A literature search revealed various measuring approaches, all of which demonstrated either little or no measurement rigour in terms of reliability. The aim of this study was to apply a robust measurement approach for testing the reliability of two-dimensional footprint impressions. Three dynamic and three static footprints were taken from the right foot of thirty female and thirty one male volunteers using the 'Inkless Shoeprint Kit'. The images were digitised. Lengths, widths and angles were measured using a selection of currently employed methods. An investigation of the reliability of the chosen measuring method suggested high intra-rater agreement: for example, the length measurement suggested an intraclass correlation coefficient (ICC) 0.99, 95% Confidence Interval (CI) -0.28 to 0.01, standard error of measurement (SEM) 0.07, Limits of Agreement (LOA) -0.91 to 0.65. Inter-rater reliability between three operators was also high: SEM ranged from 0.05 mm to 0.07 mm, ICC 0.99. Our study has established a reliable two-dimensional measuring technique that could be used for footprint comparison in further research.