(Snow Job, statistical fallacy, misunderstanding the nature of statistics [form of]; type of Half-Truth, Non Sequitur, Red Herring)

This fallacy occurs when someone deliberately supports their position using figures, numbers, and statistics that are either irrelevant or presented in a convoluted manner so as to confuse and manipulate others (different from misuse of statistics that is not deliberate). This fallacy is often mixed with other fallacies, such as overgeneralization (extrapolating to a larger group without a logical link), correlation/causation (ignoring other potential factors), and appeals to emotion.

This fallacy can appear at several stages. If the statistical test is conducted in such a way as to create a bias, such as asking loaded questions in a statistical survey, not taking random samples, or not controlling for the placebo effect, then the individual(s) conducting the study commit this fallacy. As scientific studies must be peer-reviewed and replicable, any studies that are biased are usually weeded out. Therefore, this fallacy occurs most frequently when the results of a study are then communicated to the public at large, often over-simplifying and sensationalizing the results in order to get attention. These are then further distorted by advertisers and partisan groups who then take the information to try to defend their position, often inflating, cherry-picking and distorting the actual data even further through data drudging and selectively reporting. Most people don’t recognize when this happens because the state of public statistical literacy is quite poor; human nature, based largely on intuition, is non-statistical, so most people accept studies that already agree with what they believe as opposed to forming an opinion after they have done an intensive study of it (see cognitive bias, specifically belief bias).

To guard against falling for this fallacy, demand citations for all statistical claims and check to see if the original data supports the conclusion. Avoid taking any statistical analysis by a biased party – advertisers, political groups, etc. – at face value.

Examples:

Gas prices have never been lower. When taken as a percentage of the national debt, filling up at your corner gas station is actually far cheaper today than it was in 1965!”

Did you see that bar graph in USA Today? It showed a HUGE spike in the moral decline of our country! (How do you measure morality? What is a “huge spike?” Visual representations of data can be easily manipulated.)

“Given the increasing burden of taxes on middle-class families, do you support cuts in income tax?” (as opposed to, “Considering the rising federal budget deficit and the desperate need for more revenue, do you support cuts in income tax?)

Looking at that pie chart, there is a very small percentage of people who declare themselves atheist. Therefore, atheism is not that popular of a belief. (Atheism is the lack of belief, most people can’t even define atheist, and many people identify based on culture, not religion – like Jews, for example).