After talking to the managers at the plant about your preliminary results, they explain the high costs of implementing improved safety equipment for line workers at the plant; these costs might require firing workers or reducing salaries to meet their existing budgets. The managers are willing to make worker sofety improvements if you can demonstrate with 99% certainty that a difference truly exists. You decide to use a community-based control design because you don't expect there to be enough office workers at the plant to make a precise comparison. 8. BONUS: Estimate the total sample size required for detecting a 1-unit difference in FEV1% among line workers and community controls with 90% power. Assume that there are an equal number of workers and community controls in the sample, and use the more conservative standard deviation estimate from your initial worker cohort study to estimate the population standard deviation of FEV1%. After talking to the managers at the plant about your preliminary results, they explain the high costs of implementing improved safety equipment for line workers at the plant; these costs might require firing workers or reducing salaries to meet their existing budgets. The managers are willing to make worker sofety improvements if you can demonstrate with 99% certainty that a difference truly exists. You decide to use a community-based control design because you don't expect there to be enough office workers at the plant to make a precise comparison. 8. BONUS: Estimate the total sample size required for detecting a 1-unit difference in FEV1% among line workers and community controls with 90% power. Assume that there are an equal number of workers and community controls in the sample, and use the more conservative standard deviation estimate from your initial worker cohort study to estimate the population standard deviation of FEV1%