Causal Studies - Causal Studies/Experiments Some causal...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Causal Studies/Experiments Some causal claims apply to whole populations and can be tested in the following ways: I. Controlled Cause-to Effect Experiments A random sample of a target population is randomly split into two groups: Experimental group is the group being given the suspected causal agent. Control group is the group treated the same as the experiment group but not given the suspected causal agent. C = suspected causal agent E = effect whose cause is being investigated d = the difference in the frequency of the effect in the experimental and control groups. "Statistically significant" means that there is a 95% probability that C caused E. II. NonexperimetalCause-to-Effect Study This is similar to I., but here the experimental group is not exposed to C by the investigators because it would be immoral or illegal to do so. For example, if investigators want to see if fatty diets cause cancer, the experimental group may have had extremely fatty diets while the members of the control group did not. The investigators did not give a fatty diet to members of the experimental group. The investigators then track what happens to both groups over years. III. NonexperimetalEffect-to Cause Study Start out with a control group that does not have the relevant effect (cancer of the mouth) and an experimental group that does have it. Then check to see to what extent the suspected causal agent (chewing tobacco) is present in the experimental group and not the control group. Cautions: Make sure the groups studied are large enough and that the investigators say difference is statistically significant. Avoid anecdotal evidence as it usually offers a hasty generalization. Doubtful Causal Claims 1. Circular claims -- You go to the doctor complaining of headaches on one side of your head and he tells you they are caused by "chronic paroxystic hemicrania" which means you have chronic headaches on one side of your head! 2. Nontestable claims -- there is no way to verify or falsify the claim made. 3. Excessive vagueness-- your problem is being caused by a bad attitude. What is a bad attitude? 4. Unnecessary assumptions -- I lost my car keys. Trolls must have taken them.? Go with the simpler hypothesis. 5. The explanation conflicts with a well-established theory -- theories rule! Causal Arguments Burden of proof -- which side of the dispute must give a convincing argument? In criminal law the burden is on the prosecution. In everyday arguments the burden is on the least plausible and usually the positive claim. For example, it is hard to prove one is not a racist but much easier to prove one is.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Hypothesis -- an idea or set of ideas under investigation. Confirming or disconfirming evidence --
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 5

Causal Studies - Causal Studies/Experiments Some causal...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online