epi.bohning {epiR} | R Documentation |
A test for overdispersion of Poisson data.
epi.bohning(obs, exp, alpha = 0.05)
obs |
the observed number of cases in each area. |
exp |
the expected number of cases in each area. |
alpha |
alpha level to be used for the test of significance. Must be a single number between 0 and 1. |
A data frame with two elements: test.statistic
, Bohning's test statistic and p.value
the associated P-value.
Bohning D (2000). Computer-assisted Analysis of Mixtures and Applications. Chapman and Hall, Boca Raton.
Ugarte MD, Ibanez B, Militino AF (2006). Modelling risks in disease mapping. Statistical Methods in Medical Research 15: 21 - 35.
## EXAMPLE 1: data(epi.SClip) obs <- epi.SClip$cases pop <- epi.SClip$population exp <- (sum(obs) / sum(pop)) * pop epi.bohning(obs, exp, alpha = 0.05) ## Bohning's test was used to determine if there was statistically significant ## overdispersion in lip cancer cases across 56 Scottish districts for the ## period 1975 to 1980. ## The test statistic was 53.33. The associated P value was <0.01. We reject ## the null hypothesis of no over dispersion and accept the null hypothesis ## concluding that the lip cancer data are over dispersed.