3.3.3 Study design and sample size
This study was repeated cross sectional design involving collection of rodent sample across season (wet and dry season). Preliminary survey was conducted in a study area one week before rodent sampling begun to request permission and inform authority about this study. Sherman live animal traps were used to capture rodents in farm land, bush land, peridomestic areas and forest buffer zone.
Rodent sample size was estimated with formula described by Nainget al . (2006)
\begin{equation} n=\frac{Z^{2}\text{\ P\ }\left(1-P\right)}{d^{2}}\nonumber \\ \end{equation}
Whereby n= sample size; Z= test statistics at 95% confidence level;P = expected prevalence of Yersinia pestis in rodents; (1-p ) = probability of having no infected rodent; d = desired level of precision. Values in the formula are: Z=1.96, d = 5% and P = 10%. By inserting the values into the formula\((\frac{\par \begin{matrix}{1.96}^{2}&\times 0.1(1-0.1)\\ \end{matrix}}{{0.05}^{2}}\)= 138.29). Therefore, sample size for this study was estimated to be 140 rodents in each season (wet and dry) giving a total sample size of 280 rodents.