Jowett and O'Donnell observe that 'there is a clear revival of interest in the important role of propaganda in many aspects of modern life, not necessarily related to international intrigue or military campaigns' (1992, p. xi). This thesis has examined the 1991 Census communication campaign (ABS 1991a) for evidence of propaganda intentions and strategies. Propaganda is clearly a pejorative term and its application to a particular campaign could imply a covert attempt at manipulation by the authority behind the campaign�the Australian Bureau of Statistics (ABS). As the author was responsible for the development, implementation and evaluation of the 1991 Census communication campaign and had privileged access to a range of sources, a reader might expect detailed references to internal files which reveal conscious decision-making to mount a propaganda campaign. This is not the case. An analysis of the 150 working files associated with the campaign revealed no support for such a hypothesis; there was no evidence of decision-making with the intent of implementing propaganda strategies. Similarly, during interviews with senior ABS managers, these managers categorically rejected the notion that the Bureau conducted a propaganda campaign and pointed to the fact that all procedures were cleared through both the Federal Parliament and the Privacy Commissioner. The hypothesis explored by this thesis is that despite this lack of conscious direction, propaganda processes are evident in the way the ABS conducted the communication campaign for the 1991 Census. The perspective of the thesis is closely aligned to that of Altheide and Johnson who locate propaganda as the bridge between 'organisational image and reality' (1980, p. 4). Altheide and Johnson regard propaganda as an insidious phenomena based on impression-management through the 'rigorous pursuit of scientifically valid procedures and standards' (1980, p. 229). The end result of this impression-management is that certain 'facts' are presented to the exclusion of all others. This thesis argues that impression-management strategies are evident in the way the ABS conducted the communication campaign for the 1991 Census. The processes of impression-management are subtle and do not reside in such sources as internal files. The process operates through the internalised ethos and corporate values inculcated in the minds of senior staff within the Bureau and is best conceptualised as a mindset, reflected in outcomes. I have used the term mindset to cover the process of converting abstract values into specific guides for action�fora discussion of this process see Hall (1977, pp. 69 - 83). This mindset is well-illustrated by the issue of compulsion�the obligation to complete a census form. At one level the ABS procedures are impeccable: cleared through the federal parliament and the Privacy Commissioner�and it is this form of discourse that is documented in internal files. The procedures do not, however, enable respondents to make an informed decision about whether the census is compulsory and about the ramifications for non-compliance. The mindset operating here is based on the value of the census to the ABS�the census is good for the ABS�it generates revenue and legitimises the role of the organisation. The thesis presents data which establishes that there is a significant gap between the organisational image of the census (in the corporate mind of the ABS) and that perceived by householders. The mindset of the ABS is clearly evident in the procedures adopted on this issue. The main finding of this thesis is that many of the processes underlying the development of the 1991 Census communication campaign were subtle environmental influences. These reflected the internal dynamics of the ABS, and its ability as an institution to control the communication environment through addressing the needs of other major organisations, such as the media, and the release of selective information to specific target audiences. In this context, institutional dynamics, more than decisions by individual managers, influenced the conduct of the campaign. The process of propaganda, therefore, is implicit in the 1991 Census rather than explicit�a distinction, in terms of process, drawn by Pearlin and Rosenberg (1954) in their examination of propaganda techniques in institutional advertising. It should also be acknowledged that whilst the author did have privileged access to information, there is no information contained in this thesis that would not be publicly available. The majority of primary sources are reports published by the ABS or papers delivered by Bureau staff at a range of fora. Permission was sought from the ABS for assistance in obtaining access to information and this was readily granted.