The role of waste data in changing behaviour : the case of the South African waste information system (SAWIS)

Abstract

The South African waste information system (SAWIS) developed and implemented by the Department of Environmental Affairs in 2006, provided a unique case study to explore the research question “Can the collection of data for a national waste information system change the way waste is managed in South Africa, such that there is a noticeable improvement?” The research adopted an inductive approach, incrementally constructing a conceptual model of the knowledgeable, situated waste actor, through observation and hypothesis-building and -testing. The thesis draws on theory from the fields of environmental information disclosure, science communication, environmental education, and environmental psychology, in an effort to understand and contextualise the influence of waste data and knowledge on waste behaviour. Two empirical studies were undertaken in 2006 and 2011. The studies explored whether SAWIS could create opportunities beyond simply being a tool for data collection, by building the waste knowledge of those persons tasked with the responsibility of collecting and reporting the data. The thesis posited that this new knowledge could lead to changes in personal behaviour and ultimately changes in the way organisations manage their waste. While Miller & Morris’ (1999) theoretical framework of learning provided a useful means of interpreting the 2006 data, the results showed the theoretical framework to be overly simplistic for understanding the role of waste data in a developing country context such as South Africa, in that it did not account for all of the evidence gathered, particularly the existence of behavioural and situational influences. The preliminary theoretical framework was expanded in the 2011 empirical study by including Ajzen’s (1985) theory of planned behaviour. Situated within a pragmatic paradigm, the research adopted a mixed-methods research approach, making use of both quantitative and qualitative methods. The results showed that of the three constructs of knowledge (experience, data, and theory), experience currently has the greatest influence on building waste knowledge, nearly twice that of data/information and three times that of theory. Together the three variables (experience, data, and theory) account for 54.1% of the variance in waste knowledge. Knowledge is shown to have a significant influence on all three of the antecedents to behavioural intention – attitude, subjective norm and perceived behavioural control. Furthermore, perceived behavioural control, and not intention, has the greatest influence on waste behaviour, with the model accounting for 53.7% of the variance in behaviour. Respondents from public and private waste organisations represent two distinct sub-groups in the data set, subject to significantly different influences and behaviours, creating two local models. The theoretical framework accounts for 47.8% of the variance in behaviour in the municipal local model, and 57.6% of the variance in behaviour in the private local model. By applying the combined learning-behaviour theories, the results showed that there are only three regressors that currently have a significant effect on waste behaviour, viz experience, knowledge and perceived behavioural control. Two important conclusions were reached by combining the learning-behaviour theories. First, that there are obstacles that hinder the translation of intention into behaviour in the South African context, which suggests that good waste management practice is not always under the volitional control of those tasked with its implementation. Second, that there are significant differences in the way waste knowledge and behaviour are constructed, which suggest that there are underlying social forces that shape waste behaviour and that these forces may be different in public and private waste organisations. Recognising the influence of both societal structures and agency, the theoretical framework was further expanded by embedding the two linear learning-behaviour theories within Giddens’ (1984) theory of structuration. The conceptual model of the knowledgeable, situated actor developed through this research, provides a means of understanding these barriers to action and the societal context within which waste management takes place in South Africa. From the results it is clear that a tension exists between the national neo-liberal, capitalist economic structures which support a pro-growth paradigm, and the political structures which support a pro-poor social paradigm. Furthermore, this tension plays out within a country undergoing political and organisational transformation post-1994. These structures directly influence the way waste is managed. This research proposes that by understanding the way in which knowledge and behaviour are constructed, and the societal context within which this takes place, it is possible to identify practical interventions that will lead to an improvement in the way waste is managed in South Africa.