9DESCRIBE the scope of inference that is appropriate

9EVALUATE whether a statistical study has been carried out in an ethical manner.

Lesson 4E ~ Using studies wisely OBJECTIVES:

What type of inference can be made from a particular study? The answer depends on the design of the study.

Note that in this chart, the inference is only made about the population or cause and effect. THOSE ARE THE ONLY TYPES OF INFERENCE WE CAN MAKE. Additionally, the type of inference we can make lies solely in looking at how the individuals were chosen to participate.

Well-designed experiments randomly assign individuals to treatment groups. However, most experiments don’t select experimental units at random from the larger population. That limits such experiments to inference about cause and effect. Observational studies don’t randomly assign individuals to groups, which rules out inference about cause and effect. Observational studies that use random sampling can make inferences about the population.

Example: Many students claim that the study better when listening to music, but a teacher doubts this claim and suspects that listening to music actually hurts academic performance. We will look at 4 different study designs to address the question. In each design the response variable is the students’ GPA at the end of the semester. For each design, assume that the mean GPA for the students who listen to music while studying was significantly lower than the mean GPA of students who didn’t listen to music while studying.  What can we conclude from each design?

1. Get all the students in your AP Statistics class to participate in a study. Ask them whether or not they study with music on and divide them into two groups based on their answer to this question.

2. Select a random sample of students from your school to participate in a study. Then divide them into two groups as in Design 1.

3. Get all the students in your AP Stats class to participate in a study. Randomly assign half of the students to listen to music while studying for the entire semester and have the remaining half abstain from listening to music while studying.

4. Select a random sample of students from your school to participate in a study. Randomly assign half of the students to listen to music while studying for the entire semester and have the remaining half abstain.

The Challenges of Establishing Causation: A well-designed experiment tells us that changes in the explanatory variable cause changes in the response variable. Lack of realism can limit our ability to apply the conclusions of an experiment to the settings of greatest interest.

Example: Does including the number of calories on a menu encourage diners to make healthier choices? Two AP Stats students designed a study to find out. They randomly assigned students to look at one of two different menus and indicate what they would order if they dined at the restaurant. One of the menus listed the number of calories for each item, and the other one didn’t.  Their result showed that there was no significant difference in the average number of calories ordered. How could a lack of realism affect the outcome of this study?

In some cases it isn’t practical or ethical to do an experiment. Consider these questions: •Does texting while driving increase the risk of having an accident? •Does going to church regularly help people live longer? •Does smoking cause lung cancer?

It is sometimes possible to build a strong case for causation in the absence of experiments by considering data from observational studies.

Example: Do tanning beds cause skin cancer? Doctors have noticed that people who frequently visit tanning salons are at a much greater risk for skin cancer. But are the tanning beds the cause? Could the association be due to another variable such as sun exposure? What would be wrong about designing an experiment to test this question?

If we can’t do an experiment, we can use the following criteria for establishing causation.

� The association is strong. � The association is consistent. � Larger values of the explanatory variable are associated

with stronger responses. � The alleged cause precedes the effect in time. � The alleged cause is plausible.

Consider the Surgeon General’s warning that smoking causes lung cancer.  It wouldn’t be ethical to force people to smoke just to see if they get lung cancer, but do we have enough evidence from other sources?

Complex issues of data ethics arise when we collect data from people. Here are some basic standards of data ethics that must be obeyed by all studies that gather data from human subjects, both observational studies and experiments.

Homework:

Page 273: #97-105, 113, 114