64.9k views
4 votes
A researcher wishes to estimate, with 99% confidence, the population proportion of likely U.S. voters who think Congress is doing a good or excellent job. Her estimate must be accurate within 4% of the true proportion. (a) No preliminary estimate is available. Find the minimum sample size needed. (b) Find the minimum sample size needed, using a prior study that found that 28% of the respondents said they think Congress is doing a good or excellent job. (c) Compare the results from parts (a) and (b). (a) What is the minimum sample size needed assuming that no prior information is available?

User Chronium
by
7.7k points

1 Answer

1 vote

Final answer:

To estimate the minimum sample size for a 99% confidence level and 4% margin of error without prior information, we would need 1,041 participants. Using a prior estimate of 28%, we would need a sample size of 760 participants. The sample size is larger when no prior estimate is available due to maximum variance assumption at a proportion of 0.5.

Step-by-step explanation:

To estimate the minimum sample size needed for a population proportion with 99% confidence and a margin of error of 4%, the standard formula for sample size estimation is used:



n = (Z^2 * p * (1 - p)) / E^2



Where:

  • n is the sample size
  • Z is the Z-score corresponding to the desired level of confidence
  • p is the estimated proportion of the population
  • E is the margin of error



For part a, with no preliminary estimate available, we use p = 0.5 because this gives the maximum sample size. The Z-score for a 99% confidence interval is approximately 2.576.



n = (2.576^2 * 0.5 * 0.5) / 0.04^2

This calculation results in a sample size of approximately 1,040.96, which we would round up to 1,041 participants as we can't have a fraction of a participant.



For part b, using a prior study estimate of 28% (p = 0.28):



n = (2.576^2 * 0.28 * (1 - 0.28)) / 0.04^2



This results in a sample size of approximately 759.69, rounded up to 760 participants.



In part c, when comparing the results from parts a and b, the initial estimate without prior information required a larger sample size than when a preliminary estimate was available. This is because the highest variance of a proportion occurs at 0.5, leading to the largest sample size requirement when no prior estimate is used.