David was measuring the amount of time he spends watching TV every day. To analyze his habits, for 14 days David recorded the amount of time he spent watching TV. He determines an average of 30 minutes with a standard deviation of 3 minutes was spent.

He used the same data set to construct a 90%

confidence interval to estimate the population mean time, but he found that the interval is useless as it is too wide. What should David do with his analysis in order to correct this problem?