Scenario for Questions 24 to 26, continued 
 

26.

Most of the women in this sample felt that their actual weight was 
a.

about the same as their ideal weight. 

b.

less than their ideal weight.  

c.

greater than their ideal weight. 

d.

no more than 2 pounds different from their ideal weight. 

 
27.

A chi-square test of the relationship between personal perception of emotional health and 
marital status led to rejection of the null hypothesis, indicating that there is a relationship 
between these two variables.  One conclusion that can be drawn is: 
a.

Marriage leads to better emotional health. 

b.

Better emotional health leads to marriage. 

c.

The more emotionally healthy someone is, the more likely they are to be married. 

d.

There are likely to be confounding variables related to both emotional health and 
marital status. 

 
28.

A chi-square test involves a set of counts called “expected counts.” What are the expected 
counts? 
a.

Hypothetical counts that would occur of the alternative hypothesis were true. 

b.

Hypothetical counts that would occur if the null hypothesis were true. 

c.

The actual counts that did occur in the observed data. 

d.

The long-run counts that would be expected if the observed counts are representative. 

 
29.

Pick the choice that best completes the following sentence.  If a relationship between two 
variables is called statistically significant, it means the investigators think the variables are 

a.

related in the population represented by the sample. 

b.

not related in the population represented by the sample. 

c.

related in the sample due to chance alone. 

d.

very important. 

 
30.

Simpson's Paradox occurs when 
a.

No baseline risk is given, so it is not know whether or not a high relative risk has 
practical importance. 

b.

A confounding variable rather than the explanatory variable is responsible for a change in 
the response variable. 

c.

The direction of the relationship between two variables changes when the categories of 
a confounding variable are taken into account. 

d.

The results of a test are statistically significant but are really due to chance.