The Effect of Horsfall-Barratt Category Size on the Accuracy and Reliability of Estimates of Pecan Scab Severity

Abstract

Bock, C. H., Wood, B. W., van den Bosch, F., Parnell, S., and Gottwald, T. R. 2013. The effect of Horsfall-Barratt category size on the accuracy and reliability of estimates of pecan scab sevedty. Plant Dis. 97;797-806.Pecan scab (Pusicladium effusum) is a destructive pecan disease. Disease assessments may be made using interval-scale-based methods or estimates of sevedty to the nearest percent area diseased. To explore the effects of rating method-Horsfall-Barratt (H-B) scale estimates versus nearest percent estimates (NPEs)-on the accuracy and reliability of sevedty estimates over different actual pecan scab sevedty ranges on fruit valves, raters assessed two cohorts of images with actual area (0 to 6, 6+ to 25%, and 25+ to 75%) diseased. Mean estimated disease within each actual disease sevedty range varied substantially. Means estimated by NPE within each actual disease sevedty range were not necessarily good predictors of the H-B scale estimate at <25% sevedty. H-B estimates by raters most often placed severity in the wrong category compared with actual disease. Measures of bias, accuracy, precision, and agreement using Lin's concordance correlation depended on the range of actual sevedty, with improvements increasing with actual disease sevedty category (from 0 to 6 through 25* to 75%); however, the improvement was unaffected by the H-B assessments. Bootstrap analysis indicated that NPEs provided either equally good or more accurate and precise estimate of disease compared with the H-B scale at sevedties of 25+ to 75%. Inter-rater reliability using NPEs was greater at 25+ to 75% actual disease sevedty compared with using the H-B scale. Using NPEs compared with the H-B scale will more often result in more precise and accurate estimates of pecan scab sevedty, particularly when estimating actual disease sevedties of 25+ to 75%.

A comparison between visual estimates and image analysis measurements to determine septoria leaf blotch severity in winter wheat

“…No previous study has subdivided the 0–100% ratio scale into percentage subdivisions (0–10, 10–20 … 90–100%) to address differences in agreement with the actual values over small distances of the percentage scale, although Bock et al . () did compare class groupings of a category scale with actual values. In subdividing the scale, comparisons are made over relatively small severity ranges (10% sections), allowing them to be characterized and compared.…”

“…These approaches are used in the field to improve and maintain rater accuracy. Furthermore, it has also provided the basis for avoiding certain methods that are less accurate (Nutter & Esker, ; Bock et al ., , ). The characteristics of rater error described in this study confirms previous observations and also demonstrates a tendency of precision to decline at severity >80%, and for raters to most often underestimate severity at 90–100%.…”

Effects of rater bias and assessment method on disease severity estimation with regard to hypothesis testing

“…There is a growing body of literature comparing different methods for estimating disease severity based on empirical studies (Forbes & Korva, ; Nita et al ., ; Bock et al ., ,b, , ,b; Bardsley & Ngugi, ). Although Forbes & Korva () were the first to use simulation to study aspects of disease assessment, only recently has simulation modelling of disease assessment been used to understand the impacts on hypothesis testing, thereby providing a basis to compare the assessment methods quantitatively (Bock et al ., ; Chiang et al ., ).…”

Biology, epidemiology and control of Fusicladium eriobotryae, causal agent of loquat scab

“…Similar results were obtained for V. nashicola (Umemoto, 1990). Gradients in disease were reported for pecan scab in pecan tree canopies, indicating splash dispersal (Bock et al, 2013b), and the distribution of lesions of peach scab (caused by F. carpophilum) on peach fruit was also indicative of a splash dispersed pathogen (Bock et al, 2011b).…”