Colleagues, We would like to alert you to our new paper on measurement error: "Interrogating Random and Systematic Measurement Error in Morphometric Data", currently In Press at Evolutionary Biology (https://link.springer.com/article/10.1007/s11692-024-09627-6: also available upon request).
Some of you may recall the lively discussion on Morphmet last year regarding the evaluation of ME. This paper is our 'analytical followup' to that discussion. The paper provides a thorough overview of the subject as it pertains to morphometric data, and identifies several key issues with its evaluation. First, the commonly used intra-class correlation coefficient (ICC) can be misleading, simply because a greater disparity in shape among observations can inflate ICC estimates. This could lead one to conclude that ME does not introduce concerns for the analysis of shape variation. We demonstrate that reliance on ICC statistics (or R^2 statistics) should be avoided, especially in lieu of explicit hypothesis tests. Second, we provide a framework for parsing systematic and random components of ME, and performing hypothesis tests for systematic ME. This hypothesis testing framework can include group-specific evaluation of systematic ME, if data are sampled from different groups that could vary in shape. We also show via simulation experiments that a consistent digitizing prejudice does not necessarily yield consistent systematic ME among groups, emphasizing the importance of an analytical framework that can detect and pinpoint where issues from digitizing could arise. Additionally, we provide graphical tools that assist in interpreting which research subjects might be implicated by ME. We use these statistical and graphical tools to re-analyze a recent example, identifying some cases that could elucidate digitizing issues (and could be revisited). All of these tools (both statistical and graphical) are available in the latest versions of geomorph and RRPP, now found on CRAN. Dean Adams and Mike Collyer Dr. Dean C. Adams Distinguished Professor of Evolutionary Biology Department of Ecology, Evolution, and Organismal Biology Iowa State University https://faculty.sites.iastate.edu/dcadams/ phone: 515-294-3834 -- You received this message because you are subscribed to the Google Groups "Morphmet" group. To unsubscribe from this group and stop receiving emails from it, send an email to morphmet2+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/morphmet2/CO6PR04MB842703E2EA533DB9E2F23AFAA2212%40CO6PR04MB8427.namprd04.prod.outlook.com.
