Conjoint analysis as a tool for communication and corporate environmental decision-making

Sammanfattning: Conjoint analysis has been used extensively in marketing, transportation and healthcare for the past 50 years. Since the mid-1990s, it has also been applied to an increasing number of environmental issues. In this thesis, conjoint analysis is applied to environmental objectives of the Swedish steel industry. The aim of the research project has been to develop a method for eliciting stakeholder preferences for environmental issues and to present the results to the stakeholders through the multivariate data analysis method of partial least squares regression.This licentiate thesis is based on three studies: a literature review of all conjoint studies conducted on environmental issues, a pilot study in which a new method was developed, and a main study where this method was tested on four environmental objectives of the Swedish steel industry. Up to the present, 109 environmental conjoint studies have been carried out. Recently, studies in environ¬mental economy have dominated, with investigators trying to estimate a mone¬tary value on environmental values or products. The problem selection has varied widely, from ecosystem management and green products to waste and risk management.One trend among environmental conjoint studies has been to use conjoint methods that report results at a group level. In the pilot study here, conjoint analysis was combined with partial least squares regression to enable presentation of results at individual level. The benefit of individual-level results is that they can be used to initiate a dialogue between an industry and its stakeholders, as well as between different stakeholder groups. The presentation mode also allows the researcher to find segments of respondents that think alike, although perhaps from different groups. In the pilot study, cluster analysis was used successfully to find hidden segments among the respondents.In the main study, the method from the pilot study was used to elicit preferences for four environmental objectives of the Swedish steel industry: reduced use of non-renewable energy, reduced use of non-renewable resources, decreased emission of carbon dioxide and weight reduction of products. Six stakeholder groups participated in the study, and the results showed that decreased emission of carbon dioxide was the environmental objective that most participants prioritised highest. Lowest priority was given to weight reduction of products.Conjoint analysis combined with multivariate data analysis methods such as partial least squares regression, principal component analysis and cluster analysis has been verified as an effective method for eliciting stake¬holder preferences on industry-wide environmental issues.