In: Statistics and Probability
A program director for a large computer maker is evaluating a circuit-designing software package. Not only is the program purported to ease the engineer’s tasks, but its author also claims that the finished products will be more economical because fewer logical elements will be needed. To test this theory, the software was borrowed and used to redesign 20 prototype circuits already completed the traditional way. The resulting computer-generated designs all worked and involved a mean percentage reduction in components of 3.4% (S = 4.2%). a) At the 1% significance level, can the project manager reject the hypothesis that the software design involved at least as many elements as do traditional ones? b) What is the smallest significance level at which the null hypothesis can be rejected?