6.19 Establishing Causation
Baby Einstein makes your kid smarter?
http://video.google.com/videoplay?docid=3888709234236729342#
"No Einstein in your Crib. Get a Refund." (New York Times, October 24, 2009)
http://www.nytimes.com/2009/10/24/education/24baby.html?_r=1&hp
Once we have conducted our experiment, we may have established that changes in one
variable (explanatory) have a strong association with the changes in another variable
(response).
There are three possible reasons behind this.
a. Causation - Changes in x causes changes in y.
b. Common Response - Changes in x and y are caused by lurking variable z.
c. Confounding - any effect of x on y is confounded by the effect of lurking variable z on y.
A change in the explanatory variable ( ), correlates with a change in the response variable ( )
a. Causation
b. Common Response
6. Experimental Design I Page 1
c. Confounding
6.20 Exercise - Causation?
Which example could be explained by causation, common response or
confounding? If you explain the example by common response or
confounding, identify a third variable that causes this.
1. People who are religious and go to a place of worship regularly live longer than those who
do not.
2. A high GPA causes you to have a high SAT score.
3. Does more education cause a higher income?
4. A group of girls aged 9 to 12 who had their body mass index (BMI), a measure of
weight relative to height was studied. For each subject the number of hours per week
of physical activity, hours of weekly television viewing, amount and type of food
consumed and BMI of their mothers was also recorded. The strongest correlation was
between the BMI of the girls and their mothers.
Watch the video about
correlation (okay, it's a
PowerPoint - but it's good!)
https://castatistics.wikispaces.com/file/view/
We+have+correlation.ppsx
1.
2.
3.
4.
People who take attend church or mosque or synagogue also take better care of themselves than non attenders. They are also le ss likely to smoke,
more likely to exercise, and less likely to be overweight. The effects of these good habits are confounded with the direct ef fects of attending
religious services.
The positive correlation is explained by this common response to the lurking variables of students' ability and knowledge.
However, confounding is also present. People who have high ability and come from prosperous homes are more likely to get many years of
education than people who are less able or poorer. Of course, people who start out able and rich are more likely to have high earnings even without
much education. We can't say how much of the higher income of well -educated people is actually caused by their education.
Body type is in part determined by heredity. Daughters inherit half their genes from their mothers. As a result there is a di rect causal link between
the BMI of mothers and their daughters. However, even when causation is present, it is rarely a complete explanation of an association between
two variables. It may well be that mothers who are overweight also set an example of little exercise, poor eating habits, and lots of television. Their
daughters pick up these habits to some extent, so the influence of heredity is mixed up with the influences from the girls' e nvironment. We call this
mixing of influences confounding.
6. Experimental Design I Page 1