blog #2: Correlation Is Not Causation

One of the fundamental principles of sociology is that every theory has to be backed by research and evidence. A successful social science research focuses on observing and documenting identifiable, repeating patterns of human though and action. When empirical evidence suggests that there is a correlation between variable X and variable Y, it does not automatically mean that X causes Y to happen (Hoynes 36).

The essay association is not causation addresses the fallacy that alcohol and drugs cause violence. This fallacy comes from the theory that alcohol and drugs makes causes people to release their inhibitions thereby causing violent behavior. In the essay Gelles and Cavanaugh share the arguments and evidence against the theory that alcohol and drugs causes violence: Cross-cultural evidence shows that there is variation in drinking behavior based on what a particular culture believes about alcohol (Gelles 2), this cross-cultural evidence was put to laboratory test which further proved that drinking was related to aggression only as a function of expectancy; Blood tests of men arrested for wife beating showed that only 20% of the men were legally intoxicated, while a national survey showed strong links between alcohol and violent behavior, analysis of drinking behavior at the time of the violent incident clearly demonstrates that alcohol was not used immediately prior to the violent conflict in the majority (76 percent) of the cases (3). All this researches did not document pre-use personalities, which would support a causal relationship.

The essay highlights major problems that can happen when collecting and analyzing data. In studies linking alcohol and drugs to violence, the concepts are not clearly and universally defined, for example the terms family violence, domestic violence, violence, intimate violence, and abuse are often used interchangeably. This raises the question of validity of the data measured and compromises conclusions. An analysis of the research designs of the links between alcohol and violent behavior shows that most research designs do not have control groups, which helps determine whether significant correlation exists; data collection happens at one point in time which makes it hard to determine whether there is in effect causality between intoxication and violence. The key evidence against the disinhibition theory however is the absence of a theoretical rationale. This means that research undermines the theory that alcohol and some drugs chemically affect the brain and break down or reduce inhibitions, and thus cause violent behavior.