We have a 7 point scale to measure the construct: likelihood of buying a new car. We wish to estimate the mean score on the scale and we want to be 95% confident that our sample statistic (mean) is within +/- 0.5 of a scales point from the true population parameter.
A. find out the required sample size.
B. After you find out the sample size in 27 a, you decide you want to halve the lever of accuracy. What would the new sample size be?