The entire collection of individuals or objects about which ...

[APUS57]

The entire collection of individuals or objects about which information is desired is called a sample.
   
  1.True
  2.False

Question 2

Suppose the researcher described in 2 wants to test for differences in the ratings of well-being before and after commencing the gluten-free diet. Why might a test for paired differences using the Student's T distribution not be appropriate?
 A. The two measures of well-being are independent.
  B. The two measures of well-being are not quantitative.
  C. The distribution of differences may not be normally distributed.
  D. The difference between the two measures of well-being cannot be computed.

[Anton]

Answer to Question 1

2

Answer to Question 2

C.The distribution of differences may not be normally distributed.

FEEDBACK:
Since the ratings of well-being are measured on a 1 to 5 scale, they are quantitative and differences can be computed. However, since the sample size is small, it's hard to imagine that those differences could be normally distributed. Take a moment to think about what a histogram of those differences might look like. Perhaps make up 10 people and their ratings and then draw a histogram of the differences in those ratings. Such a distribution could take on many shapes, but with a maximum difference of 5 - 1 = 4, the shape is unlikely to be mound-shaped and/or symmetrical, like the normal distribution. While much of the statistics you have learned in this course depends on the data being normally distributed, realize that this necessary assumption is often not the case, particularly when the sample size is small. In this example, the distribution of differences are probably not normally distributed such that you cannot trust that a test using the Student T distribution would produce a valid conclusion.

[APUS57]

YES! Correct, THANKS for helping me on my review

[mochi09]

TYSM