Health insurance companies look for ways to lower costs, and one way is to shorten the length of hospital stays. A study to compare two large insurance companies on length of stay (LOS) for paediatric asthma patients randomly sampled 393 cases from Insurer A. Summary statistics and a histogram of the data are shown in Display 9.37.
a. Is it appropriate to construct a confidence interval without transforming these data? Explain.
Â b. Regardless of your answer to part a, estimate the mean LOS for Insurer A in a 90% confidence interval.
c. Would you be more concerned about constructing a confidence interval without a transformation if the sample size was 40 instead of nearly 400? What about a sample size of 4 instead of nearly 400?