The thesis investigates the effects that the Research Assessment Exercise (RAE) has on the Higher Education sector. The alternative view presented by the thesis is that more knowledge can be created by concentrating on the different constituents of the RAE and their specific interactions with particular areas of the Higher Education sector. The RAE constituents are interpreted as drivers that influence and impact, in dissimilar fashions, on different activities of Higher Education Institutions (HEI). This micro analysis of the RAE enables the investigation to isolate the single effects of the RAE drivers therefore creating a bottom-up analysis of the overall impact of the RAE. The analysis of the impact that the drivers have on HEIs’ activities focuses on the perception that individuals within the system have of the consequences of the RAE. The focus on perceptions derives from personal observation of the lack of consensus on the consequences that different drivers have on different areas. The use of perceptions as the mean to assess the impact of the RAE enables the investigation to create a picture of the consequences of the RAE that addresses behavioural change. A multi-dimensional crystal view approach is used to accommodate both the micro analysis and the perception assessment. The multi-dimensional crystal view, a research contribution in its own right, is based on the principle that a micro analysis of a complex system can be achieved by decomposing the system into a number of dimensions. Insight is draw when the interactions between some of the dimensions are investigated. In the specific case of the RAE the dimension are: the RAE drivers, HEIs’ activities and points of observation (dimension that captures perceptions). Knowledge and insight can be acquired when the interactions between the dimensions are aggregated at successive higher levels. The supporting tool for the multidimensional crystal view approach is a matrix that facilitates the analytical process. The aggregation of the dimensions comes from combining textual statements from the points of observation (perceptions) on the effects that the drivers of the RAE have on the activities of HEIs. The highest level is a textual statement that synthesises all lower level statements.

Cette thèse étudie les effets de deux instruments d’action publique, le Research Assessment Exercise (RAE) et le full Economic Costing (fEC), sur le mode de gestion du travail des universitaires anglais. À partir de quatre études de cas —quatre départements, repartis sur deux universités et deux disciplines, géographie et informatique—, cette thèse montre que les deux instruments étudiés ont des effets structurants sur l’organisation formelle des départements.Les deux instruments introduisent le langage financier et le calcul monétaire dans les relations des différents acteurs, imposent le développement d’outils de gestion et, plus largement, participent à la managérialisation des universités. Le RAE et le fEC, grâce au travail d’interprétation fait par les universitaires, simplifient les demandes de l’environnement vis-à-vis des départements. Ceci permet la mise en place d’outils de gestion dont l’objectif n’est pas seulement de légitimer le département vis-à-vis des évaluateurs externes, mais également de coordonner le travail des universitaires. Cependant, le RAE et le fEC ont surtout des effets indirects. Le mode d’organisation des départements n’est pas déterminé par les instruments nationaux, mais par des négociations qui dépendent de la capacité des universitaires à rapporter des ressources matérielles et symboliques à l’université. Il peut être managériale collégiale dans les départements qui obtiennent de bons résultats financiers et au RAE ou managériale hiérarchique dans ceux qui ont du mal à satisfaire les attentes de la direction de l’université. Dans les deux cas, les universitaires gardent un degré d'autonomie. / In this thesis we study the effects of two policy instruments, the Research Assessment Exercise (RAE) and the full Economic Costing (fEC), on the management and the work of academics on English departments. From four case studies of four departments in two universities and two disciplines —Geography and Informatics—, this thesis shows that the two instruments studied have a structural effect on the formal organization of departments. Both instruments introduce financial language and monetary calculation in the relations of actors, they require the development of management tools and more broadly, they are involved in the managerialisation of universities. The RAE and the fEC, through the work of interpretation by the academics, simplify the demands of the environment vis-à-vis the departments. This allows the implementation of management tools whose objective are not only to legitimize the department vis-à-vis the external evaluators, but also to coordinate the work of academics. However, the RAE and the fEC have effects mainly in an indirect way. Instruments are not only interpreted at the departmental level, but they are also operationalized through management tools that open new spaces of academic freedom. The department organization is not determined by national instruments, but by negotiations that depend on the ability of academics to bring material and symbolic resources to the university. It can be collegial-managerial in departments that get good funding and good results in the RAE or hierarchical-managerial in those that are struggling to meet the expectations of the management of the university. In both cases, the academics retain a degree of autonomy.