It has long been acknowledged that changing attitudes and norms can be a successful route to adjusting unsustainable consumption habits in the developed world. What is less clear is how communication campaigns work towards encouraging these changes. This paper details how combining constructs from two well-respected theories, one from the field of social cognition and another from the field of communication study, could deliver enhanced communication effectiveness for those in civil society charged with ensuring the long term viability of their communities. Data were gathered from participants engaged in a range of activities aimed to change environmentally significant behaviours. These activities were organised either by a nonprofit environmental action group which developed out of the Agenda 21 programme, or by the local government body for the area. Both organisations belong to a strategic partnership of individuals and groups whose purpose is to help everyone working or living in their community to achieve a one tonne per capita reduction in the carbon dioxide emissions by 2010/11. The activities were directed at adjusting behaviours associated with consumption in the fields of domestic energy use, food sourcing, travel behaviour or waste. The aim was to investigate the relationship between variables from Petty and Caccioppo’s Elaboration Likelihood Model (ELM), such as perceptions about a message and its source, and the variables proposed by Ajzen’s Theory of Planned Behaviour (TPB) as being key to behaviour change, such as subjective norms and attitudes. From a communication professional perspective, the ELM adds vital specification to the TPB in explaining behaviour. While communicators need to be mindful of psychological factors influencing the targets of their communication when they devise programmes, they more usually have direct influence over external variables such as message and source. Two questionnaires were used to capture variable constructs for both theories. These were drawn as much as possible from survey items found to have high reliability in previously published research. The first questionnaire was conducted at the time participants took part in an activity. The second was conducted between four and five weeks later and included questions about behaviours since activity. Analysis was conducted using standard multiple regression to test the strength of relationships between the predictor variables listed earlier and the outcome variable of behaviour change. Moderation and mediation analysis were used to investigate the relationships between the predictor variables. This paper reports the findings of this analysis and concludes that the impact of communication and social cognition predictor variables, along with their capacity to modify or mediate the impact of other predictors, provides communications designers with a checklist of issues to consider, and guidance for their relative importance, when designing environmentally significant behaviour change communications.