Simply go to the Ghana Statistical Service, they can provide you with the data on the total in each enumeration area. And by the way are you sampling from all the enumeration area. It does not appear possible to sample from the entire enumeration area so u may have to either purposively decide or randomly select from the 31. If it becomes diffract to get the total number in each enumeration area, you can do the addition by requesting the towns that make up each enumeration area and add up the randomly sampled population. In that case you arrive at your target population and you may apply the single population proportion formula provided u know of a previously reported prevalence of mob justice:

You can borrow insignts from Israel 1992 sample size table and also by

Morgan, D. W., & Krejcie, R. V. (1970). Determining sample size for research activities. Educational and psychological measurement, 30(3), 607-10.

Hope this helps

If all the variables within the research units all matters, then you can randomly select the respondents. If not, you can use a non-probability purposive sampling technique. Since the population is a big one, you need to use the following formula:

n = N over 1+N (e) exponent 2. See below:

📷

Where n = sample size

N = population

e = margin of error

(Source: Jones & Gomez, 2010).

This will help you arrive at the sample size for the research.

Secondly, to obtain a sample size for each of the 31 communities, you need to consider the following formula:

P = (population of community over total number of the communities) x n. See below:

P 📷