Table 2 Definitions of mechanism (adapted from Mahoney [14])

Definition

An intervening (set of) variable(s) that explain(s) why a correlation exists between an independent and dependent variable

Frequently occurring causal patterns that are triggered under generally unknown conditions and with indeterminate consequences. A mechanism explains by opening up the black box and showing the cogs and wheels of the internal machinery. It provides a continuous and contiguous chain of causal or intentional links between the explanans and the explanandum

An unobserved entity that, when activated, generates an outcome of interest

Analytical approach

Correlational analysis techniques, such as mediation analysis, are used to identify “mechanisms” that are considered to be mediators of the observed effect

While slightly more broadly defined, this definition is compatible with probabilistic approaches to analysis

Causal analysis consists of identifying the configuration that links the outcome to mechanism(s) triggered by the context, often combining quantitative and qualitative data

Role given to theory

Theories in the form of universal laws can be deduced from empirical research (covering law principle)

Theories in the form of empirical knowledge derived from constant conjunction observations

Research contributes to developing theories of the middle range

Implications

Risk of reduction of mechanisms to measurable indicators, through which dynamic processes of change are reduced to correlations between variables that stand for more complex processes

In this view, and similar to definition 1, causation is reduced to the concatenation of elements in a causal chain. Causation is demonstrated to the degree that empirical regularities can identified

Empirical research allows investigation of a possible mechanism, thus identifying a plausible mechanism and may eventually lead to the identification of the actual mechanism. Research thus contributes to increasing the plausibility of the explanatory hypothesis