Skip to main content
Fig. 2 | International Journal of Health Geographics

Fig. 2

From: Distance sampling for epidemiology: an interactive tool for estimating under-reporting of cases from clinic data

Fig. 2

Confusion matrices for classification obtained with the multinomial models of reported malaria cases at 8 health centres (on x and y axes), as a function of distance from village of origin, in Komoé district, Burkina Faso. The diagonal elements represent the percentage of cases for which the predicted classification (case being assigned to a given clinic) is equal to the true classification, while off-diagonal elements are those that are mislabelled by the classifier. The predicted cases are obtained using a a model considering only the effect of distance, b a model considering the effect of distance and age, c a model considering the effect of distance and sex, d a model considering the effect of distance and season

Back to article page