Final answer:
The criterion for deciding when a statistical result is not random error is the alpha level, which has to be set at a probability of 0.01.
Step-by-step explanation:
The criterion for deciding when a statistical result is not random error is the alpha level, which has to be set at a probability of 0.01. This means that the probability of observing the result by chance alone should be less than 0.01 in order to conclude that it is not due to random error.
For example, if a study finds that the average number of doctor visits for women is 5.8 times per year, and the alpha level is set at 0.01, there would need to be strong evidence to conclude that women visit their doctors more than 5.8 times per year.