Question

Bags of a certain brand of tortilla chips claim to have a net weight of 14 ounces. Net weights actually vary slightly from bag to bag and are normally distributed with mean u. A representative of a consumer advocate group wishes to see if there is any evidence that the mean net weight is less than advertised and so intends to test the hypotheses Ho: u = 14, Hai u < 14. To do this, he selects 16 bags of this brand at random and determines the net weight of each. He finds the sample mean to be x = 13.88 and the sample standard deviation to be s-0.24. Reference: Ref 7-2 Suppose we were not sure if the distribution of net weights was normal. In which of the following circumstances would we not be safe using a t procedure in this problem? The mean and median of the data are nearly equal. O A histogram of the data shows moderate skewness. A stemplot of the data has a large outlier. The sample standard deviation is large.

Answer 1

Option(C): A stemplot of the data has a large outlier.

>> A large outlier can inflate the sample variability, decreasing the t statistic and thus perhaps eliminating a significant difference. A nonparametric test may be a more powerful test in such situation.