A new tool is supposed to reduce the time it takes to assemble a component in a factory. A random sample of workers is selected, and the assembly times are measured with and without the new tool. The average difference in assembly time is 3.4 minutes, with a standard deviation of 4.6 minutes. Sample size is 18.You may assume that the differences are normally distributed. Is there evidence, at the 5% level of significance, that the new tool speeds up assembly time? The order of subtraction is (time without tool - time with tool).