Unit 2b Review: Misleading Statistics and Causation vs. Correlation

Vocabulary flashcards covering methods of misleading statistics, graphical distortions, and causal fallacies from the Unit 2b Review.

Correlation

A relationship between any two variables, where a change in one is seen alongside a change in another, but which does not mean that one variable causes the other.

Sample Selection

A method of misleading readers through statistics by choosing a non-representative or biased group of volunteers, such as polling members of a specific political party to predict general election results.

Detached Statistic

A statistical claim that leaves out a point of comparison, such as stating that 'More than $80\%$ of dentists recommend Colgate' without clarifying what the toothpaste is being compared to.

Ambiguous Averages

One of the seven methods by which statistics can be used to mislead readers, involving unclear or unspecified measures of central tendency.

Faulty Survey Questions

One of the seven ways readers can be misled through statistics, occurring when survey questions are poorly designed, biased, or confusing.

Implied Connection

One of the seven ways readers can be misled via statistics, subtly suggesting a connection or cause-and-effect relationship without stating it outright.

Change the Subject

A tactic where statistics are used misleadingly by shifting the focus away from the original issue or what is actually being measured.

Omitting the Baseline

A technique used in misleading graphs where the vertical axis does not start at zero, exaggerating differences between values.

Manipulating the y-axis

A method of creating a misleading graph by distorting scales, omitting units, or presenting no scale at all to influence the reader's perception.

Cherry Picking

One of the five ways graphs can be misleading, involving the selective presentation of only data that supports a specific conclusion while ignoring contrary data.

Using Wrong Graph

A misleading graphical method where an inappropriate graph type is selected to misrepresent the underlying data.

Going Against Conventions

One of the five ways graphs can mislead readers, occurring when standard graphical norms (such as scale directions or layouts) are violated to confuse the viewer.

Reverse Fallacy

A causal fallacy where the belief is that $X \rightarrow Y$, but in reality, $Y \rightarrow X$.

Common Cause Fallacy

A causal fallacy where the belief is that $X \rightarrow Y$, but in reality, an outside variable $Z$ causes both ($Z \rightarrow X$ and $Z \rightarrow Y$).

Bidirectional Fallacy

A causal fallacy where the belief is that $X \rightarrow Y$, but in reality, both variables cause each other ($X \rightarrow Y$ and $Y \rightarrow X$).

Oversimplification Fallacy

A causal fallacy where the belief is that one variable causes another ($X \rightarrow Y$), but in reality, multiple factors contribute to the outcome ($A, B, C, D, \text{ and } E, X \rightarrow Y$).

Coincidence Fallacy

A causal fallacy where two correlated variables are believed to have a causal relationship ($X \rightarrow Y$), but in reality, they just so happen to occur together without causation.