Correlation Does Not Prove Causation.

¶ … correlation does not prove causation.

According to Hersen and Thomas (2003), "Correlation does not prove causation'" is a statement every aspiring psychologist should learn. "A study may find a negative correlation between depression and self-esteem such that people with lower self-esteem are found to report higher levels of depression," they advise, and, "The temptation is to conclude that people are depressed because they have low self-esteem (and that by raising self-esteem, depression will be reduced)" (Hersen and Thomas 15). Researchers, though, must not succumb to this temptation to jump to a conclusion because there is nothing in the data that supports such a causal inference. In many cases, it is reasonable to assume that something is caused by something else, but appearances can be misleading and it is can be easy to confuse reality with external factors that may not be recognized or acknowledged as contributing to the observed outcome. In this regard, a good example that illustrates why correlation and causation are not always related can be found in the following vignette.

A man and a dog are sitting in a room facing each other. When the man touches his nose, he observes the dog begin to salivate, become agitated and wag its tail. When the man removes his finger, the dog whines. Based on these observations, the man concludes that the dog is happy when he touches his nose and unhappy when he stops. The man writes his findings and publishes them in a scholarly journal, but receives an enormous amount of criticism because of his specious findings. It was revealed by another researcher that what the man did not realize was that there was always a woman standing behind him with a juicy steak that the dog could see but he could not. When the man touched his nose, the woman showed the steak to the dog and the dog reacted to the steak. When the man removed his finger, the woman hid the steak and the dog whined. The man involved was devastated by his erroneous jump to an inaccurate conclusion. Thereafter, he never wrote again and routinely kicked his dog.

According to Glenberg (1996), the fact that correlation does not prove causation "should be etched indelibly into your brain (right next to 'You can't prove the null hypothesis'). Demonstrating a relationship, either through correlational analysis or through regression analysis, does not establish a causal link between the two populations. Causal links can only be established using experimental methodology" (487). For this purpose, the primary factor needed in such an experiment involves gathering data in two conditions that are exactly the same (or at least have no systematic differences) except for the level of the independent variable (Glenberg 487). This author adds that, "One condition has one level of the independent variable and the other condition has a different level. Because the conditions have no systematic differences except for the level of the independent variable, any statistically significant difference on the dependent variable must be caused by the independent variable (or a Type I error)" (Glenberg 487).

Furthermore, Hersen and Thomas (2003) emphasize that in some cases, a third variable can influence the causal relationship between the other two and in other cases two or more factors cause each other. Nevertheless, despite these constraints, Goldstein and Strube (1995) report that while, "correlation does not prove causation, but we acknowledge that the first step in inferring causality is often the detection of a correlation" (205).

2. Explain the arguments both favoring and opposing the use of animals in research.

If presented with the choice between having some type of odious experiment conducted on themselves or animals, even the most ardent antivivisectionists would likely vote for animals but there are some profound ethical issues involved in this decision. For example, Varner (1998) points out that, "Animals are not ours to eat, wear, or experiment on' is a politically expedient slogan for someone who believes that radical reform is called for, even if that person actually believes that some uses are acceptable under some conditions" (99).

Those in strict favor of using animals in research will point to the important advances in medicine that have been achieved over the years that have saved countless human lives. By contrast, strict opponents of using animals in any type of research suggest that animals have rights that preclude their being used in this fashion, not matter how important the scientific findings that might result from their use. For example, according to Kistler (2002):

As for animal research, I believe it is satanic. I realize that sounds melodramatic, but I don't use that word lightly. Some things like concentration camps and modern warfare are satanic. Laboratories are places that are as fiendish as slaughterhouses. What makes the problem worse in the case of animal research is that the people involved are presumably intellectuals -- intellectuals without moral concern or moral awareness. The animal research laboratory is an excellent example of the derangement and fragmentation of modern intellectual life -- the split between intellect and ethics. (161)

As noted above, others adopt a more middle-of-the-road position that says it is morally acceptable to use animals in some types of research, but not in others. According to Galvin and Herzog (1992), "The use of animals as subjects in behavioral and biomedical research has become a major issue with social, political, philosophical, and psychological ramifications. Largely due to increasing pressure from both animal activists and the public at large, decisions concerning the treatment of experimental animals are no longer routine" (263). These authors add that this is not a straightforward moral analysis, but requires some careful scrutiny to understand the respective positions of the advocates and opponents of using animals in research.

Despite the assertions by Galvin and Herzog (1992) noted above, in reality, it would seem that this is a real "out of sight, out of mind" issue for most people today and it is reasonable to assume that unless the issue of animal rights is brought up, many consumers will not think about how animals were used in the development of their cosmetics or textiles, but will simply wear them. Likewise, most people will not stop to think about the conditions animals are subjected to during their processing when they purchase a Big Mac or a bucket of chicken at KFC. Nevertheless, these animals are also subjected to conditions that might shock many people, so it appears that the arguments both in favor of and opposed to animals being used in research require much more public education to become relevant issues in the future. In this regard, Galvin and Herzog conclude that:

The complexity of this type of decision-making process illustrates the diversity of attitudes regarding the use of animals in research. Understanding the processes individuals bring to bear on moral dilemmas involving the use of animals may open avenues of communication among those voicing opposing views on the animal research issue. Although perfect consensus may not be achievable, understanding and compromise may be within reach. (285)

3. After taking a bite of food, describe the role each part of the brain plays in this "simple" act.