D'après le magazine Terra eco (2014),48% des français pratiquent régulièrement la consommation collaborative et 32% ont l'intention de le faire. Malgré l'engouement des consommateurs pour ces alternatives d'échanges, nous constatons comparativement assez peu d'études sur ce sujet en marketing. Cette recherche se propose donc d'analyser les motivations des membres des réseaux physiques de consommation collaborative, à travers l'étude du Système d'échange local (SEL). Le chapitre I propose un état des lieux du SEL, à travers son histoire, son étendue, son objet et son fonctionnement. Le chapitre 2 présente le SEL sous son approche académique. Par une revue de la littérature, il montre comment les différentes disciplines expliquent cette réalité. Il se penche plus précisément sur les déterminants au sélisme, et se termine par la proposition d'un modèle d'influence des motivations au sélisme et de leurs antécédents sur la pratique séliste. Le chapitre 3, par une analyse de l6 entretiens semi-directifs de sélistes, complète et enrichi le modèle. Le chapitre 4traite de la méthodologie employée pour le tester, Le chapitre 5 s'attèle à la création d'une échelle des motivations au sélisme, ainsi qu'à la validation des échelles de mesure du modèle. Le chapitre 6 se consacre enfin au test du modèle et à I'approfondissement des résultats. Cette étude quantitative se base sur un échantillon de 957 sélistes. Les principaux apports de notre thèse consistent : en une modélisation du comportement du séliste ; en la découverte de deux nouvelles motivations à faire partie d'un SEL : se distraire et s'enrichir intellectuellement; en la création d'une échelle de mesure des motivations au sélisme; en l'analyse de leur influence sur la pratique séliste ; et en la création d'une typologie de sélistes. Pour finir et sur la base de ces résultats, des préconisations managériales sont proposées aux SEL, aux entrepreneurs sociaux ou privés de réseau de consommation collaborative, ainsi qu'à l'État et aux collectivités territoriales. / According to the magazine Terra Eco (20 I 4), 48%o of French people are regular users of collaborative consumption, and 32%o of them are considering doing it. Despite the consumer craze for those alternative exchanges, comparatively, only a few marketing studies on the topic are currently made, This research's aim is therefore to analyse the various motivations of members from a physical network of collaborative consumption through the study of the local exchange trading system (LETS). The first chapter is focused on a state of play of the LETS, through its history, its geographical extent, its aims and uses. The second chapter will observe the LETS under an academic point of view. Through a litterature review, it will show how the differents disciplines are explaining this reality. It will also highlight in details what is determining LETS practices, and will end up suggesting an influence model on motivations behind becoming a LETS member, and their antecedents to the LETS practice. The third chapter will strenghten the model by a deep analysis of l6 semi-directives interviews of LETS users. The fourth chapter will put a light on the methodology used to test it. The fifth chapter will define a motivation scale on why becoming a LETS member, and to validate the measuring scales of the model. The sixth chapter will finally test the model and dig the results deeper. This quantitative study is based on a sample of 957 LETS users. Our thesis main assets are: a modelisation of the LETS member behavior, the discovery of two new motivations to join a LETS-distraction and intellectual enrichment, the creation of a motivation scale behind the decision of joining a LETS, analysis of their influences on the LETS practice, and the creation of a LETS member typology. To conclude, and based on those results, we'll come up with managing advices to give to LETS, to social or private workers from collaborative consumption networks, and also to the State and to territorial collectivities.

This study is an in-depth investigation of the factors that affect store brand purchases. It aims to help both retailers and manufacturers predict store brand purchases through an improved understanding of the effects of three latent variables: customer satisfaction and loyalty with the store; which is expressed through word-of-mouth; and trust in store brands. An additional aim is to explore variations in the level of store brand adoption and the inter-relationships between the selected constructs. Data was collected through a telephone survey of those responsible for household grocery shopping, and who shop at the nine leading grocery retailers in Greece. A total of 904 respondents completed the questionnaire based upon a quota of 100 respondents for each of the nine retailers. Data were analyzed through chi-square, analysis of variance and partial least square. The proposed model was tested by partial least square path modeling, which related the latent variables to the dependent manifest variable: store brand purchases. The findings provide empirical support that store brand purchases are positively influenced by the consumers’ perceived level of trust in store brands. The consumer decision-making process for store brands is complex and establishing customer satisfaction and loyalty with the store does not appear to influence store brand purchases or the level of trust in the retailer’s store brands in the specific context under study. Consequently the most appropriate way to influence store brand purchases in the Greek market is through increasing in the level of trust in the retailer’s store brands. It is suggested that retailers should therefore invest in trust building strategies for their own store brands and try to capitalize on their brand equity by using a family brand policy. Theoretical and managerial implications of the findings are discussed and opportunities for further research are suggested.

This study consisted of two distinct research phases, performed within the context of the South African Council for Science and Industrial Research (and its current and potential research alliances). The purpose of the study was to develop a decision making model that would enable strategists at publicly financed research and development organisations to analyse and predict governance mode decisions, as well as select optimal governance mode structures (ranging from quasi-market structures, such as once-off contracts, to quasi-hierarchy structures, such as research joint ventures) for research alliances. During the qualitative first phase, the study aimed to identify impact domains within South Africa’s new Bayh-Dole-like Intellectual Property Rights legislative framework that consists of the Intellectual Property Rights from Publicly Financed Research and Development Act, as well as the Technology Innovation Agency Act, which could potentially influence research alliances (based on the Transactional Cost Economics, Resource-based View and Real Options Approach perspectives) with publicly financed research and development organisations. This was followed by the quantitative second phase, which attempted to verify the validity of a value-mediated governance mode model that included the highest ranked impact domains identified during the first phase as formative indicators for the perceived Intellectual Property Rights regime strength uncertainty factor. A qualitative online survey amongst senior managers at the Council for Science and Industrial Research, followed by Theme Extraction combined with Constant Comparative Method analysis, as well as a weighted frequency analysis, constituted the research methodology employed during the first phase’s identification and ranking of impact domains within the South African legislative framework. This phase demonstrated that the highest ranked impact domains (primarily driven by the Transactional Cost Economics perspective) included the choice of Intellectual Property Rights ownership, state walk-in rights on undeclared Intellectual Property, and benefit-sharing policies for the creators of Intellectual Property. The second phase consisted of a quantitative online survey, distributed amongst current and potential research alliance partners of the Council for Science and Industrial Research, followed by Structural Equation Modelling of a value-mediated governance model that included, amongst others, the perceived Intellectual Property Rights regime strength as an uncertainty factor. This phase revealed not only that the impact domains identified during the first phase could be used as formative indicators of the perceived Intellectual Property Rights regime strength, but also that stronger perceived regimes are positively related to the preference for quasi-hierarchy research alliance governance modes. Furthermore, it established that the expected value of a research alliance, which was shown to be positively influenced by the strength of the perceived Intellectual Property Rights regime, acted as a mediating factor on the relationship between the perceived Intellectual Property Rights regime strength and the preferred research alliance governance mode. Keywords: Bayh-Dole, Formative Indicators, Intellectual Property Rights, Research Alliances, Real Options Approach, Resource-based View, Quasi-Market Governance Modes, Quasi- Hierarchy Governance Modes, Structural Equation Modelling, Transactional Cost Economics, Value-mediated Governance Model. Copyright / Dissertation (MBA)--University of Pretoria, 2012. / Gordon Institute of Business Science (GIBS) / unrestricted