Where does association collide?

The measurable tendency of two variables to vary together, so that knowing the value of one changes the probability distribution of the other: an association is any statistical dependence, quantified by measures like the odds ratio, the relative risk, or a correlation coefficient, and it holds whether or not one variable actually causes the other. Austin Bradford Hill's 1965 The Environment and Disease: Association or Causation? laid out the reasoning by which an observed association might, or might not, be judged causal, and stressed that dependence alone never proves cause. The referent is a numerical relationship between measured variables, and it is not a link between ideas in a mind, a community of plants, or an organised group of people.[1]

Unit

A dependence between two variables, scaled by an effect measure such as an odds ratio

Recovery means

Not applicable

Measured by

Quantified by odds ratios, relative risks, and correlation coefficients, and tested against the hypothesis of independence

A learned connection between two mental representations, so that activating one tends to bring the other to mind: the tradition of associationism, running from Locke and Hume into experimental psychology, holds that experience binds ideas, stimuli, and responses together whenever they occur close in time or resemble one another. David Shanks's 2010 review Learning: From Association to Cognition traced how such associative learning is studied and where a purely associative account meets its limits. The referent is a bond inside an individual mind, revealed by how one item cues another, and it is not a statistical dependence in data, a plant community, or a chartered group of members.[2]

Unit

A learned bond between two representations, graded by how strongly one cues the other

Recovery means

The bond weakens through extinction when the paired items stop occurring together

Measured by

Assessed by cued recall, reaction time, priming, and conditioning procedures that gauge how one item evokes another

A recurring, characteristic assemblage of plant species that tends to occur together under similar conditions, treated as a nameable unit of vegetation: phytosociology, developed by Clements in North America and Braun-Blanquet in Europe, defined the association as the basic type into which plant communities are classified. Robert Whittaker's 1962 monograph Classification of Natural Communities surveyed these competing schemes and the long argument over whether such units are real entities or convenient abstractions. The referent is a repeatable grouping of species in a landscape, mapped and named by its typical plants, and it has nothing to do with a bond between ideas, a dependence in data, or a club of people.[3]

Unit

A named vegetation type defined by its characteristic and diagnostic plant species

Recovery means

Not applicable

Measured by

Identified by sampling species composition and abundance in plots and grouping stands by their diagnostic species

A voluntary organisation that people form and join to pursue a shared purpose, from a choir or bowling league to a professional body or civic group: such associations are a core part of civil society, and since Tocqueville they have been read as schools of cooperation that build trust and civic capacity. Robert Putnam's 1995 Bowling Alone: America's Declining Social Capital argued that falling membership in these groups was eroding the stock of social capital in the United States. The referent is an organised body of people bound by membership and shared aims, and it is not a statistical dependence, a mental bond, or a plant community.[4]

Unit

An organised membership group, counted by members, meetings, and the density of such groups

Recovery means

Not applicable

Measured by

Gauged by membership rolls, participation rates, and survey measures of joining and civic engagement