If pilot test revealed 45% perceived poor service quality, and we choose the interval range is + 10 percent with a 95% confidence level, what is the computed sample size to confirm or not the findings from pilot test previously applied to firm's customers?