This thesis examines the construct of Psychological Sense of Community (PSOC). Within the discipline of community psychology, there is debate as to the dimensions underlying the construct PSOC. One of the few theoretically proposed structures is that put forward by McMillan and Chavis (1986), who hypothesized four dimensions: Belonging; Fulfilment of Needs; Influence; and Shared Connections underlying PSOC. Further, there is some deliberation in the literature as to the existence of PSOC in relational, as well as geographical, communities. Discussion has also emerged regarding the role of social identification within PSOC. It has been suggested that differences in PSOC may be understood in terms of the degree to which members identify with their community (Fisher & Sonn, 1999). However, few studies have explored the place of identification in PSOC. In addition, while PSOC has been applied to both relational and geographical communities, little research has looked in depth at PSOC within relational communities. Thus, the principle aims of the current program of research were to elucidate the underlying dimensions of PSOC and their consistency across geographical and relational communities. Further, the research also aimed to explore the role of identification in PSOC. The first stage of this research endeavoured to clarify the underlying dimensions of PSOC by utilising a questionnaire which included multiple measures of PSOC and social identification, administered to both relational and geographical community members. The first paper of the current research explored PSOC in a relational community, science fiction fandom (N = 359) and the third paper in a sample of residents of rural, regional and urban geographical communities (N = 669). In both the relational and geographical communities, support emerged for McMillan and Chavis' (1986) four dimensions of PSOC. In regards to identification, the Sense of Community iv affective and ingroups ties aspects of social identification were subsumed within the PSOC dimensions; however, the Conscious Identification aspect emerged as separate to the existing PSOC dimensions. The study presented in paper three also examined the role of demographic factors in predicting PSOC in geographical communities. The demographic factors significantly associated with PSOC were: type of region, with rural participants displaying higher PSOC than their urban counterparts; participation in local organizations; having children; and a vision of one's neighbourhood as broader than just a street or block. To date, little research has compared a single group's PSOC with a relational community to their PSOC with their geographical communities. The second paper presented in this manuscript explored PSOC with participants' relational and geographical communities in the sample of members of science fiction fandom (N = 359). All the PSOC dimensions and Conscious Identification emerged as significant predictors of overall sense of community in both community types. Participants reported higher levels of global PSOC with fandom than with their geographical communities, a pattern that also emerged across the four dimensions and Conscious Identification. It was proposed that the degree of choice of community membership may be one reason for this finding. However, stronger conclusions could not be drawn from this study as situational salience may have influenced the results as data was collected in the relational community context. The second phase of the current research aimed to validate the multidimensional nature and related measures of both social identification and PSOC. The fourth paper presented in this thesis examined the construct validity of the three-factor model of social identification as measured by the Three Dimensional Strength of Identification Scale proposed by Cameron (1999, 2004). The 12 item version of the scale was used to collect data from an undergraduate sample (N = 219) to assess their social identification across three distinct group memberships (sex, student and interest group). This data was subjected to confirmatory factor analysis to examine the fit of the three-factor model of social identity in comparison to fit indices for one and two-factor models. The results indicate that the three-factor model was the most parsimonious and best fit to the data across all groups. In addition, the fact that different patterns of means and correlations emerged across groups on the three dimensions provided further evidence for a multidimensional model of social identification and, moreover, the greater depth of exploration it allows. The fifth paper examines The Sense of Community Index (SCI), one of the most commonly used measures of PSOC. There is much discussion in the literature as to the validity of the scale as a measure not only of overall PSOC, but of the dimensions (Membership, Influence, Needs fulfillment and Emotional Connection) theorized by McMillan and Chavis (1986) to underlie the construct. This paper examines the factor structure of the Sense of Community Index in a study (N = 219)that examined neighborhood, student and interest group communities. The results showed that the Sense of Community Index, in terms of its original factor structure, did not adequately fit the data. The scale was revised, utilizing confirmatory factor analysis indicators, to produce a new four-factor structure based on the original items. This revised model was tested and found to display adequate fit indices to the data in all three community types. The results of the study provide empirical support for retaining measures that encapsulate the four dimensions of PSOC.The sixth paper further explores the interplay between PSOC and the dimensions of social identification. In particular, the study (N = 219) examines the relative strength of the separate aspects of social identification (based on Cameron's 2004, Three Factor Model of Social Identification) as predictors of overall PSOC, accounting for situational salience. Results indicate that Ingroup Ties is consistently the strongest predictor of PSOC and that the strength of Ingroup Affect and Centrality alter according to the group or community context. The seventh and final paper from the current research program emerged from the results of paper two indicating that choice may influence individuals' social identification and PSOC with their respective communities. The study presented in this paper examined participants' (N = 219) level of social identification and PSOC across multiple group memberships that differ in the degree of choice associated with membership (low choice: neighborhood community; medium choice: student community; and high choice: self chosen interest group). Results indicated that, controlling for contextual salience, choice was positively associated with levels of social identification and PSOC. Overall, the current program of research provides some important findings which add significantly to the theoretical understanding of PSOC in today's society. The research provides clarification of both the dimensions underlying PSOC, their application to both geographical and relational communities and the measurement of overall PSOC and these dimensions. Further, it provides empirical evidence of the importance of the Centrality aspect of identification in PSOC in both geographical and relational settings. The theoretical and practical implications of the findings of the overall program of research are discussed.