Fig. 2
From: Impact of offering cycle training in schools upon cycling behaviour: a natural experimental study
Conceptual models guiding analyses. MCS = Millennium Cohort Study. In this model, square boxes denote measured variables; circles denote unmeasured variables; and solid lines show hypothesised causal relationships. Part A shows the conceptual model guiding our primary analyses, concerning the effect of offering Bikeability in school upon cycling frequency. Part B shows a more detailed version of this conceptual model that hypothesises the role of participation in cycle training as a mediator, and also shows how participation in cycle training might be subject to confounding by measured and unmeasured characteristics