STAT 1350 Lecture Notes - Lecture 23: Null Hypothesis, Triiodothyronine, Statistical Hypothesis Testing
27 Apr 2018
School
Department
Course
Professor
4/5/2018
Chapter 22: More Practice and Hypothesis Testing
● Top Hat from Tuesday
○ You conduct a hypothesis test with a significance level of 0.05. You obtain a p-
value of 0.35. What should your decision be?
■ A.You accept the null hypothesis
■ B. You fail to reject the null hypothesis
■ C. You reject the alternative hypothesis
■ D. All of the above are equivalent statements
● Remember this?
○ A television ad claimed a study “proves that Puppy Chow, then Dog Chow
Nutrition, fed properly over a lifetime can help extend his lovable antics up to 1.8
healthy years.”
○ The actual study was published in the Journal of the American Veterinary
Medicine Association (2002).
○ Here is an excerpt from the study:
■ “Compared with control dogs, food-restricted dogs weighed less and had
lower body fat content and lower serum triglycerides, triiodothyronine,
insulin, and glucose concentrations. Median life span was significantly
longer for dogs in which food was restricted.”
● What does “significant” mean?
○ It depends.
○ If we say our findings are statistically significant, it means that chance alone is
unlikely to be the explanation. If we obtain a small p-value and reject the null
hypothesis, we have statistical significance.
○ If we say our findings are practically significant, it means they have important
implications, and the observed differences are not trivial. Judging practical
significance can be more subjective.
● Statistical Significance vs. Practical Importance
○ A finding can be statistically significant without being practically important.
Document Summary
You conduct a hypothesis test with a significance level of 0. 05. D. all of the above are equivalent statements. A television ad claimed a study proves that puppy chow, then dog chow. Nutrition, fed properly over a lifetime can help extend his lovable antics up to 1. 8 healthy years. The actual study was published in the journal of the american veterinary. Here is an excerpt from the study: Compared with control dogs, food-restricted dogs weighed less and had lower body fat content and lower serum triglycerides, triiodothyronine, insulin, and glucose concentrations. Median life span was significantly longer for dogs in which food was restricted. If we say our findings are statistically significant, it means that chance alone is unlikely to be the explanation. If we obtain a small p-value and reject the null hypothesis, we have statistical significance.
