 |
 |
||||
|
 |
||||
|
|
|
||||
|
|
||||
|
Using the Challenger Disaster to Help Explain Correlation
By Rob McEntarffer
Southeast High School
Lincoln, NE
Concept:
- For most students, the explosion of the space shuttle Challenger is a distinct moment from their childhood. The decision to launch Challenger was in part based on correlational analysis of failure rates and temperature. Psychology students can look at the actual data available to the experts who decided to launch the shuttle and decide themselves if they would launch the shuttle or not. This activity will demonstrate the importance of accurately interpreting correlational data.
Materials Needed:
- Data Tables A and B (See Below)
Instructions:
Start with a discussion about the explosion. Ask students where they were, what they remember thinking about the disaster, etc. Then ask them if they know what caused the disaster. You will get a variety of responses. Discuss the real reason (o-ring failure caused by low temperatures at time of launch). Explain that NASA tested the o-rings at various temperatures to determine if there was a correlation.
Show them Data Table A and ask them to draw a scatter plot of the data. Ask the students to draw the line of best fit and determine what type of correlation the graph shows: positive, negative, or no correlation. (They should say there is no correlation.)
Explain that this is why NASA launched. According to the data used, there was no correlation between temperature and failure rate. But, the problem is that they did not include all of the data. They did not include points where there was not a failure.
Show the class Data Table B and ask them to do another scatter plot. Ask the students to draw the line of best fit and determine what type of correlation there is: positive, negative, or no correlation. (They should say there is a negative correlation.) Discussion:
- Explain that the shuttle was launched in part to using data from Table A instead of Table B. Discuss the importance of using all available data in experimentation and analysis. You can also raise social pressures on NASA decision makers (group-think, etc.) to launch the shuttle during a time of budget pressures. If your statistics instruction includes how to calculate Pearson's R, students could calculate R for each table.
Data Table A
Temperature
# of Failures
53
2
57
1
58
1
63
1
70
2
75
2
Data Table B
Temperature
# of Failures
53
2
57
1
58
1
63
1
66
0
67
0
68
0
69
0
70
0
70
2
72
0
73
0
75
0
75
2
76
0
79
0
81
0
References and additional reading:
Feynman, R.P. (1988). What do you care what other people think? New York, NY: Bantam Books.
The Mathematics Teacher, 87 (6), September 1994, 423-426.
Wainer, H. (1997). Visual revelations. New York, NY: Springer-Verlag, 51-53.
The above activity was originally published in the Septermber/October 1997 issue of The Psychology Teacher Network. The activity is reprinted here with the permission of the Education Directorate of the APA. Further publication of the activity is not permitted without the express written consent of the Education Directorate.
For information on subscribing to The Psychology Teacher Network, write to:
- Psychology Teacher Network
Education Directorate
APA
750 First Street, NE,
Washington, DC 20002-4242.
