Correlation and Causation

Vocabulary flashcards covering the distinction between inductive reasoning, correlation, causation, and empirical data analysis from the lecture.

Inductive Argument

A probabilistic argument involving factual claims about the world that are contingently true rather than necessarily true.

Contingent Truth

A factual claim about the way the world happens to be that is true, but could have been different than it is.

Necessary Truth

A claim that must be true and about which we can be certain of its truth (e.g., 'if someone is a bachelor then they are an unmarried man').

Causal Argument

An inductive, probabilistic argument in support of a causal explanation or claim asserting that a particular effect has a specific cause.

Correlation

A connection or association between two kinds of things where the frequency of occurrence within a data set goes up or down together.

Hume's Account of Causation

David Hume's theory that causation is merely a habit of the mind formed when the regular experience of $A$ being followed by $B$ creates an expectation of $B$ whenever $A$ occurs.

Causation Standards (Dhawan and Shay)

Evidentiary criteria summarized by Ravi Dhawan and Denys Shay requiring temporal consistency (cause precedes effect), a strong predictable correlation, a clear biological or social explanatory mechanism, and consistency across studies.

Operationalization

The precise method by which a concept or variable is defined, observed, and measured within research studies and reporting.

Positive Correlation

A relationship in data where the values of variable $A$ tend to increase and decrease in synchrony (parallel) with the values of variable $B$.

Negative Correlation

A relationship in data where variable $A$ tends to increase when variable $B$ decreases, and vice versa.

No Correlation

A condition in data where variable $A$ changes haphazardly as variable $B$ changes, meaning the variables move independently of one another.

Haidt's Social Media Hypothesis

Jonathan Haidt's argument in 'The Anxious Generation' that the sharp rise in adverse adolescent mental health outcomes around 2012–2013 was directly caused by the expansion of social media.

Alternative Explanations to Haidt's Hypothesis

Correlational factors for rising teen mental health data starting around 2012, including destigmatization, the Affordable Care Act (2010) expanding coverage, and 2011 USPSTF guidelines mandating depression screening for teen girls.