In statistics and data analysis, causalation (a portmanteau of the words "causation" and "correlation") is the fallacy of claiming a cause and effect relationship when either:
- none exists
- the evidence does not support such a claim
Frequently, a causal relationship can only be established after careful experimentation following the scientific method. Even then, it may be very difficult to determine the true relationship between two variables, due to the effects of lurking variables or other complicating matters.
One excellent example is the positive correlation between ice cream sales and violent crime (i.e. violent crime increases as ice cream sales increase). The fallacy of causalation would claim that ice cream sales cause violent crime (or vice versa). However, such a claim ignores lurking variables such as heat (which would drive up both crime and ice cream sales).
Causalation is often used to skew statistics in favor of a particular argument. For instance, one might observe a supposed correlation between homosexuality and mental illness and claim that being gay leads to psychiatric disorders (not that anyone makes that claim... oh, wait). What this ignores is the effect of social stigma and the oppressive nature of many societies concerning LGBT persons — which, let's face it, is not very fun, and may lead to the kind of stress that sometimes causes mental illness.