The third problem above has been corrected with some Bland Altman diagrams by adapting a linear regression that describes the age difference as a function of mean age (Giarvina 2015). However, this only allows for a linear relationship that does not constitute or reveal more interesting nonlinear relationships. A generalized additive model (GAM) could be used to estimate a potentially nonlinear “smoothed” relationship between age differences and mean ages. In situations where one of the age estimates might a priori be considered more accurate, it seems more appropriate to place this age estimate on the x-axis rather than on the mean between it and the less accurate estimate. In other words, go back to the concept, but not the precise structure, of the age distortion diagram. The GAM smoother can also be added to this diagram (Figures 5 and 6). Bangdiwala, S. I., Ana S. Haedo, Marcela L.
Natal and Andres Villaveces. The match diagram as an alternative to the recipient company`s usage line for diagnostic tests. Journal of Clinical Epidemiology, 61(9), 866-874. A function that indicates the fill colors used for exact and partial matches Weights can be indicated to allow partial matching taking into account the contributions of off-diagonal cells. Partial concordance is usually represented in the display by a lighter shade, as given by fill_col (j) corresponding to the weights [j]. The Bland Altman diagram in Figure 1 was created with bland.altman.plot () from the BlandAltmanLeh package (Lehnert 2015b). There are other R functions for creating Bland Altman plots (or the equivalent “Tukey`s Mean Difference Plot”). It is, however, a simple diagram that can be easily created from “background”, as shown next.
I then gently criticize the Bland Altman action for use in age comparisons and offer an alternative (this is not an act of age distortion). Figure 1: Bland Altman diagram to compare scale with Lake Champlain Whitefish otolith time estimates. Thiw was built with BlandAltmanLeh. Establishes a classification table from raw data in the table for two observers and calculates an inter-rater-agreement (Kappa) statistic to assess the concordance between two classifications on the ordinal or nominal scales. If you add it all up, you get the diagram in Figure 3. These results indicate that, up to an average age estimate of about five, there is little difference between the scale and age estimates of the otolith, according to which the age estimates of the scales are lower than the age estimates of the otoliths, with the difference between the two in general increasing with the average age. . . .