Abstract
This thesis is concerned with role-related differences in evaluation and their consequences. Many researchers assume that it is possible to evaluate objectively from a single perspective. An exploration of the literature concerning evaluation research and role theory however suggests that: 1. People differ in their evaluations of a given target. (e. g. Glass and Ellet, 1980; Weiss et al, 1973). 2. Differences in evaluation can occur with respect to: (i) the content of evaluations, as evidenced by the indicators of evaluation considered important, (e. g. Broady, 1968; Lynch, 1977). (ii) the level of evaluation (e. g. Wineman, 1982; Zimring and Reizenstein, 1980). 3. These differences are related in part to the roles which people occupy with respect to the target of evaluation (e. g. Sandahl, 1974). 4. Much evaluation literature distinguishes between two major types of roles, often referred to as provider roles and user roles. Two major differences between them are that: (i) providers are more likely to control the product of evaluation, and to make decisions affecting its nature and existence (e. g. Glass and Ellet, 1980; Knight and Campbell, 1980). (ii) users are likely to have more direct experience of the product, and to experience the consequences of providers' decisions (e. g. Weiss et al, 1973; Knight and Campbell, 1980). Exploration of literature concerned with the characteristics of different role-occupants (e. g. Deasy, 1974; Hershberger 1970) suggests that role-titles act as labels which summarize training and/or experiences of role-occupants. The research reported here used a specific context - a local authority housing system - in order to: (i) identify a network of providers and users with a common focus of evaluation. (ii) examine the relationship between role-related characteristics and the indicators of evaluation considered important by (a) housing providers and (b) housing users. (iii) compare the indicators of evaluation used by providers and users when evaluating the same target, and trace the consequences of the differences between them. Data was collected by means of open-ended interviews with housing providers and users. Content analysis and Smallest Space Analysis of this data provided information about the indicators of evaluation spontaneously identified by different role-occupants. This information was considered in the light of role-related characteristics clarified by multidimensional scalogram analysis. The existence of differences in content of evaluation for occupants of different roles was confirmed. A major difference was found to exist between providers' and users' evaluations of the same target of evaluation. Provider roles with respect to the target of evaluation were more varied than were user roles; and evaluations used by providers reflected their professional training and experiences. The evaluations of users were found to be related at a detailed level to the personal experiences and circumstances of users. The concepts of professional role and personal role are introduced to clarify these differences in evaluation. A model of evaluation is presented. The implications of these findings for evaluation research and for the experiences of users are considered. Evaluation researchers should not assume that evaluation from a single perspective is objective; and providers involved at all stages of the design and use of a product need to seek information about the indicators upon which users base their own evaluations.