A manufacturer of watches has established that on the average his watches do not gain or lose. He also would like to claim that at least 95% of the watches are accurate to ±0.2 s per week. A random sample of 15 watches provided the following gains (+) or losses (-) in seconds in one week:
Can the claim be made with a 5% chance of being wrong? (Assume that the inaccuracies of these watches are normally distributed.)