I am trying to determine the sample size for an ongoing study on traditional authority. I need to determine the sample size for the population of people who have dealt with traditional authority on public construction projects.
When the population is unknown, you can estimate the sample size based on population proportion.
A population proportion, generally denoted by p and in some textbooks by pi (symbol), is a parameter that describes a percentage value associated with a population. For example, the 2010 United States Census showed that 83.7% of the American Population was identified as not being Hispanic or Latino. The value of p = .837 is a population proportion.
To calculate the sample size based on the sample required to estimate a proportion with an approximate 95% confidence level , you can use the following formula:
n=((1.96^2)pq)/(d^2)
Where n = required sample size, p = proportion of the population having the characteristic, q = 1-p and d = the degree of precision. If the proportion of the population (p) is unknown use p = 0.5 which assumes maximum heterogeneity (i.e. a 50/50 split). The degree of precision (d) is the margin of error that is acceptable. Setting d = 0.02, for example, would give a margin of error of plus or minus 2%.
Basically you can conservatively estimate the proportion of the target population to give you a sample size.
When the population is unknown, you can estimate the sample size based on population proportion.
A population proportion, generally denoted by p and in some textbooks by pi (symbol), is a parameter that describes a percentage value associated with a population. For example, the 2010 United States Census showed that 83.7% of the American Population was identified as not being Hispanic or Latino. The value of p = .837 is a population proportion.
To calculate the sample size based on the sample required to estimate a proportion with an approximate 95% confidence level , you can use the following formula:
n=((1.96^2)pq)/(d^2)
Where n = required sample size, p = proportion of the population having the characteristic, q = 1-p and d = the degree of precision. If the proportion of the population (p) is unknown use p = 0.5 which assumes maximum heterogeneity (i.e. a 50/50 split). The degree of precision (d) is the margin of error that is acceptable. Setting d = 0.02, for example, would give a margin of error of plus or minus 2%.
Basically you can conservatively estimate the proportion of the target population to give you a sample size.
Once you are dealing with large populations (over 200 or 500), the effect of population size on your estimated sample size is pretty much the same all the way out to infinity.
So, as Christopher Bova points out, the variance of the variable you are trying to measure is the key issue. Also, as he indicates, if you do not have a prior estimate of "p" then assume you have a two-category variables that is divided 50-50.
The 'population' is actually defined as ALL people, including 'people who have dealt with traditional authority on public construction projects'. So you need to do a survey to determine which people in the the population meet your criteria, then use a sample of those people in your study. You can determine sample size for the survey and study using tables and online calculators that determine ideal sample size based on confidence limits, margin of error, and population size, or do a statistical power test :)
you can take previous studies and give a reference while writing manuscript.
Hi,
Beside the sample calculations aforementioned, you have to consider the size of the population you are measuring, is it in the context of institution, region ? Country ? continent ? For each context there should be a different sample size. It also depends on what your measuring, in your case it is traditional authority on public construction projects, so I think the calculations of the sample size should start from the areas that are affected mostly by these projects.
I want to distribute a questionnaire to experts who are working on smart cities project in my country. But I don't know the population of the experts and accordingly I am not sure of the sample size I have to consider. please not that smart cities are new concept to my country and they are only few of experts working on smart cities projects
Hana Albulushi it sounds like you need to use a "snowball sample" approach, and aim to get as much coverage of the total population as you can.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size