You are the assistant director of political research for the ABC television netw

Question

You are the assistant director of political research for the ABC television network, and two candidates, Joseph Thomas and Ann Martin, are running for president of the United States of America. You need to furnish a prediction of the percentage of the vote going to Martin, assuming the election was held today, for tomorrow’s evening newscast. You want to be 95% confident in your prediction and desire a total precision of 6%.

a. Assume that you have no reliable information concerning the percentage of the population that prefers Martin. What sample size will you use for the project? (2.5 points)

Type your answer below and show your calculations.

b. Assume that a similar poll, taken 30 days ago, revealed that 40% of the respondents would vote for Martin. Taking this information into account, what sample size will you use for the project? (2.5 points)

Type your answer below and show your calculations.

c. Which of the two sample sizes you have just calculated would you prefer to use for your study? Why? (2.5 points)

Type your answer below.

d. Most polls of this type are conducted by telephone. When a potential respondent answers the phone, what is the first question you should ask? Why? (2.5 points)

Please type your answer below.

Explanation / Answer

In case P represents the proportion, then Q = 1-P and the standard deviation also known as standard error in case of sampling is SQRT(PQ/n) where n is size of the sample. In order to validate the assumption that sample proportion follows normal distribution, it is necessary to have large samples. Sample size may be as per the population size but an accepted norm for large sample is to have atleast of size 30. Larger the size better the results.

The formula for sample size n = C2 *P*Q/E2 where C is constant taking the values 1, 2, 3 corresponding to confidence level of 68.3%, 95% and 99%. E represents the level of error, 6% as mentioned in the question.

Assuming value of P as .5, we get n= 4*.5*.5/.0036 = 278 for 95% confidence interval

B. It is given that P=.4, therefore n= 4*.4*.6/.0036 = 267 for 95% confidence interval

In case of 99% confidence level n = 9*.4*.6/.0036 = 600 ( C=3 in case nothing is mentioned about the level of confidence interval)

C. As mentioned earlier it is always preferred to have larger samples so that the error can be reduced or minimized.

Hence minimum size of the sample should be 278 and maximum may be 600 for more accuracy.

D. I would like to ask about the preference from the potential respondent in case he/she votes.

Related Questions

Navigate

• Browse (All)
• Browse Y
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.