To reduce bias in survey estimates, most longitudinal survey organisations, nowadays, prepare and include sets of weights in public use data files for use by analysts. Aside from correcting for non-coverage, the weights are usually designed to reflect the sample design as well as to correct for non-response error by combining design weights and non-response weight adjustments together. With regard to non-response weights, many longitudinal surveys implement similar strategies (referred to as the standard weighting approach in this thesis) to create them. This approach is based upon a weighting model where: response is defined as responding at all conducted waves; all sample members whose eligibility is unknown are assumed as eligible and the model is estimated by using generic weighting variables and all sample members for which data are available on the weighting variables. However, there are several issues in longitudinal surveys that raise concerns regarding using this approach of weighting. In particular, this thesis is concerned with three challenging issues: non-monotonic response pattern which results in a large number of combinations of waves at which sample members could respond, and hence weights that result from an approach such as the one in question, which defines response as responding at all the conducted waves may not be appropriate for the analysis of data from a wave-combination that does not include all waves; unknown eligibility over time leads to including a proportion of ineligible units in the weights' calculation (if they are assumed to be eligible as in the standard approach) which may result in biased estimates unless the actual ineligible units amongst units of unknown eligibility are excluded; and the choice of the best covariates for the weighting model which may differ considerably across different subgroups of respondents in the same sample. In the standard approach only generic weighting variables are used in the weighting model, as all sample members are used in the estimation. Meanwhile, some variables, which may not be significant in predicting response for the whole sample, could be important in predicting the response in some subgroups. In this thesis, I provide three alternative approaches (each deals with one of the raised issues) for non-response weighting. I investigate each of the proposed approaches by incorporating relevant weight adjustments, as well as weights from the standard weighting approach, in a longitudinal multivariate analysis. I test the impact of weights from each alternative approach on estimates by comparing the resultant estimates with estimates resulting from the standard approach. I use data from the British Household Panel Survey (BHPS) to carry out the investigation. The findings suggest that the standard and alternative approaches, all help similarly in reducing non-response error. However, the standard approach may fail in tackling the effect of non-response in some estimates, as it does not take into account the three raised issues in the weighting of longitudinal data. In contrast, since they deal with the three issues under investigation (separately), the alternative approaches seem to handle non-response even in estimates that are not affected by the standard weighting approach.

In recent decades social scientists have started to use qualitative creative methods1 more and more, because of epistemological and methodological developments on the one hand and demands of innovation by governmental funding agencies on the other. In my thesis I look at the research practices of social scientists who use these qualitative creative methods and answer the following main research question: How are practices and approaches from the arts (specifically visual lens-based arts, poetry, performance and narrative) negotiated in social scientific research practice? This question has been divided into the following three sub-questions: 1) How do social scientists negotiate the use of creative methods with other members of their research community? 2) How do social scientists negotiate the use of creative methods into their own research practices? 3) And how do creative methods emerge in the process? Using Lave and Wenger's approach to communities of practice (1991; Wenger, 1998) and Ingold and Hallam's (2007) conceptualisation of improvisation for my theoretical framework, I look at these practices as constantly emerging and changing, but at the same time determined by those same practices. Based on ongoing conversations with postgraduate research students, interviews with experienced researchers, participant observation at conferences and videos of my participants' presentations, I conclude that the use of creative methods is always embedded within existing research practices. When this is not the case, either participants themselves or other academics experience the creative methods as problematic or even as non-academic. In those cases boundarywork (the in- and exclusion of what is deemed academic) is performed more fiercely, making it difficult, if not impossible for creative methods to be truly innovative in the sense that it means a break with previous practices. Instead, we see small shifts in participants' academic practices and how creative methods are taken up in these practices. This means improvisation is a more apt term to describe how creative methods are making their way into social scientific research practices/into the social sciences. As such this conclusion has consequences for the way we think about learning methods, the production of knowledge, innovative methods and (inter)disciplinarity.

There is increasing emphasis on the outcomes of research in terms of its impact on wider society. However in the social sciences the ways in which research is taken up and used, discussed, shared and applied in different policy, practice and wider settings is complex. This thesis set out to investigate the ways in which social research was used by various non-academic actors, and to explore what impact it had in order to develop methods for understanding and assessing impact. The research investigated what research impact is, how it occurs, and how it might be assessed. The research was in two phases: firstly, a case study of a research partnership between a research centre and a voluntary organisation; and, secondly, the development and seeking feedback on a framework to assess impact. The care study employed two main approaches: forward-tracking - from research to policy and/or practice - and backward tracking - from policy back to research. Both phases were conducted through a practitioner-research approach, bringing experience of working with the projects involved into the heart of the research model. The study found many ways the research from the partnership had been used in different sectors by different actors. Impacts from the research were harder to identify. In cases where there were clear impacts, the actors involved had adapted research to fit the context for research use in order to create impact. Research users continued to draw on the research for many years after publication, creating further impact as new policy or practice agendas arose. The framework used a 'pathways to impact' model to develop a theory-based approach to assessing impact and to create categories for data collection. The ways in which research might impact on policy and practice are many and cannot be easily predicted. Concepts from complexity theory, particularly a focus on relationships, an understanding of context and the concept of emergence have been useful in framing the picture of impact generated from this research. Any assessment of impact from social research needs to acknowledge that many actors are involved in the process of research being taken up and used, and impact cannot be achieved from the supply side alone. Partnership research, between an academic and voluntary sector organisation, facilitated the use and impact of the research in many ways. The thesis reconceptualises ideas about how research impacts on society, suggesting the concept of 'contribution' is more accurate and useful than attribution. It also adds to the body of empirical work on the processes of impact, and in particular of the role of research partnerships in increasing impact. It suggests that process-based approaches to assessing impact that acknowledge complexity may be fruitful in developing impact assessment methodology.

The purpose of this thesis is to explore the history of covert ethnography in Britain between the 1880s and 1980. During this century, a range of academic and non-academic social researchers have used the method of covert ethnography. The starting point for this thesis is the observation that there is no adequate and sustained explanation of covert ethnography as a historical phenomenon. It is argued that the fragmented nature of the existing historiography precludes a full understanding of this important historical phenomenon. It is the intention of this thesis to bridge the gaps in the historiography, as it stands, and to promote an inclusive historical account of covert ethnography in Britain across time. Through an analysis of covert ethnographic projects undertaken in Britain between the 1880s and 1980, with particular attention being paid to the structure and language used by covert ethnographers, this thesis will locate the use of this research method in its historical context. This thesis will chart the changes and continuities over time in the use of covert ethnography and demonstrate how key forces, such as the establishment of new models of ethnographic research and the development of ethical concern regarding covertness, shaped the use of covert ethnography significantly. This thesis will contribute a more comprehensive account of covert ethnography to the existing historiography.

Un possible, impossible, la coproduction des connaissances entre science et société. La recherche est une activité sociale principalement déléguée aux chercheurs professionnels au sein des laboratoires académiques ou privés, dans une certaine mise à distance du reste de la société. A la faveur du « tournant participatif », des acteurs de la société civile sont invités aux côtés des chercheurs professionnels, par de nouveaux dispositifs de financement de la recherche, à contribuer à la production de connaissances nouvelles dans des domaines d’intérêt sociétal. Que se passe-t-il dans ces recherches collaboratives? Des connaissances scientifiques peuvent-elles y être co-construites ? Et de telles collaborations sont-elles en mesure de renouveler les modes de production de la connaissance déléguée aux chercheurs ? C’est au travers d’un regard qui postule la question de la traduction comme majeure dans l’émergence d’une possible co-construction entre acteurs et chercheurs qu’est conduite cette étude de dispositif de recherche collaborative. La traduction implique des plans multiples de l’interaction qui sont autant des manières de passer entre les registres de connaissances et les intérêts des uns et des autres que des interprétations qu’ils produisent pour leur donner du sens. Cette dimension centrale de la traduction entre des mondes scientifiques et associatifs est abordée dans une double perspective épistémologique. Une première approche issue de la sociologie de la traduction permet de saisir comment acteurs et chercheurs s’associent et relient leurs différentes identités et intérêts, ainsi que leurs savoirs et les différents objets de recherche pour réaliser des inscriptions scientifiques. Il en émerge des agencements collaboratifs dont les configurations plurielles témoignent de conditions de co-construction. Une seconde approche mobilise l’herméneutique de la traduction pour rendre compte des capacités d’interprétation et d’apprentissage qui peuvent émerger de ces collaborations de recherche. Elle place au cœur des interactions les capacités de délibération et de reconnaissance qui orientent les agencements collaboratifs vers une capacité collective de recherche.A l’issue de cette analyse, l’identification de conditions d’interaction favorisant la co-construction de connaissances entre acteurs de la société civile et chercheurs professionnels permet de réinterroger la possibilité d’inscrire, au côté des modes dominants de production scientifique, un mode de coproduction qui participe d’un renouvellement des rapports entre science et société. / A possible, impossible, the co-production of knowledge between science and society. Research is primarily a social activity delegated to professional researchers in academic and private laboratories, with a certain distancing from the rest of society. Taking advantage of the "participatory turn", actors in civil society are invited alongside professional researchers, with new research funding arrangements, to contribute to the production of new knowledge in areas of societal interest. What is happening in this collaborative research? Can scientific knowledge be co-constructed? And are such collaborations able to renew the modes of production of knowledge delegated to researchers?It is from this way of looking at things that the question is posited regarding translation as a major premise in the emergence of a possible joint construction between civil society actors and researchers who are driving this collaborative research plan. Translating involves multiple planes of interaction which are in as many ways, ways of passing between different people's different registers of knowledge and interests as they are interpretations which are produced to give these registers meaning. This central dimension of translation between scientific and associative worlds is addressed in a double epistemological perspective. A firstly sociological approach to translation captures how actors and researchers combine and link their different identities and interests and their knowledge and different research objects to achieve scientific inscriptions. Collaborative arrangements emerge from this with various configurations which demonstrate the conditions of co-construction. A second approach mobilizes the hermeneutics of translation to account for interpretation and learning abilities that can emerge from these research collaborations. It places deliberation and recognition capabilities at the heart of interactions, which guide collaborative arrangements towards a collective research capacity.Following this analysis, the identification of interaction conditions favouring the co-construction of knowledge between civil society actors and professional researchers can re-examine the possibility of including, alongside the dominant modes of scientific production, a way of co-producing which is part of a renewal of the relationship between science and society.