[ Return to X200 Home Page ]



CHAPTER 7

1. Answer in a sentence or two.

What's the difference between a proportion and a correlation?



2. In the "abuse in sport" example, there was a claim that a considerable proportion of athletes suffer abuse. Explain why a correlation might have been more interesting than this simple proportion.


3. Circle P If what's referred to is a proportion, or C if it's a correlation.

P C 95% of students prefer morning classes to classes at other times.

P C Men have a higher incidence of suicide than women do.

P C Swedes have a high rate of alcoholism, but Czechs have an even higher rate.

4. Imagine that in a random sample of 50 male students and 50 female students at your university, 28 of the males have taken at least one arts class, and 23 of the females have taken at least one arts class.

Draw a complete tree diagram representing these data.

 



Draw a different tree diagram, which shows the "other way of calculating" the same data.

 



In your sample, Is there a positive or negative correlation, or no correlation, between MALE and TOOK AT LEAST ONE ARTS CLASS? Explain what numbers reveal this.







5. Imagine that in a random sample of 50 male students and 100 female students at your university, 17 of the males smoke, and 34 females smoke.

Draw a complete tree diagram representing these data.

 



In your sample, is there a positive or negative correlation, or no correlation, between FEMALE and SMOKE? Explain what numbers reveal this.

 


CHAPTER 8

To answer the following questions, you'll need to consult this chart:

Sample Size   Margin of Error (expressed as a decimal not as a percent) 
10             ±.30
50             ±.14
100            ±.10
200            ±.07
500            ±.04
1000           ±.03

 



(The chart, as well as the following questions, assume a confidence level of .95.)

1. Suppose that you observe 200 male students and 200 female students chosen at random in your university. Suppose that I 10 of the males are wearing baseball caps, and 85 of the females.

Draw a complete tree diagram representing these figures.

 



Draw a rectangle graph (box diagram) showing the data, with margins of error.

 





Do the data have statistical significance? Explain.

 


On the basis of your graph, there's sufficient evidence to claim that, IN THE SAMPLE,

T F there's a positive correlation In the population between MALE and WEARS BASEBALL CAP

T F there's a negative correlation In the population between MALE and WEARS BASEBALL CAP

T F there's no correlation between MALE and WEARS BASEBALL CAP.

On the basis of your graph, there's sufficient evidence to claim that, IN THE POPULATION,

T F there's a positive-correlation in the population between MALE and WEARS BASEBALL CAP

T F theres a negative correlation in the population between MALE and WEARS BASEBALL CAP

T F there's no correlation between MALE and WEARS BASEBALL CAP.

Give the range for possible correlation strength In the population.



2. Suppose that you observe 500 male students and 200 female students chosen at random in your university. Suppose that 220 of the males are dangling a water bottle from a knapsack, and 110 females are.

Draw a complete tree diagram representing these figures.

 



Draw a rectangle graph showing the data, with margins of error.

 

 





Do the data have statistical significance? Explain.

 


On the basis of your graph, there's sufficient evidence to claim that, IN THE SAMPLE,
T F there's a positive correlation In the population between FEMALE and DANGLES BOTTLE

T F there's a negative correlation In the population between FEMALE and DANGLES BOTTLE

T F there's no correlation between FEMALE and DANGLES BOTTLE.


On the basis of your graph, there's sufficient evidence to claim that, IN THE POPULATION,

T F there's a positive correlation In the population between FEMALE and DANGLES BOTTLE

T F there's a negative correlation In the population between FEMALE and DANGLES BOTTLE

T F there's no correlation between FEMALE and DANGLES BOTTLE.


Give the range of possible correlation strength.

 



Under what circumstances, If any, can there be good enough evidence for no correlation between two properties? Explain.