Sample size determination and sampling technique
We used a single proportion formula with a 95 percent confidence
interval, 5% margin of error, and with assumption of 50% proportion to
calculate required samples. Then multiplied by 2 (design effects), and
finally a 10% non-response rate was added.
n= \(\frac{Z\propto^{2\ }p(1-p)}{d^{2}}\) = n=\(\frac{\left(1.96\right)^{2\ }0.5\left(1-0.5\right)}{{0.05}^{2}}\)=384, 384 x 2 = 768, by adding 10% (77) \(\approx 845\)
The study participants were divided into two groups based on where they
lived (urban and rural). Our source population was the whole population
of Harar. Our study unit consisted of all participants that were
randomly chosen from each stratum. During the data collection period,
study participants who volunteered to engage in the study were included.
Those who did not want to participate in the study were excluded from
the study.