This thesis examined the usefulness of a paradox-based approach to the study and practice of organisational change. Lewis (2000) defines paradox as contradictory yet interrelated elements that seem logical in isolation but absurd and irrational when appearing simultaneously. Given that organisational change is inherently complex and contradictory (e.g., Cameron & Quinn, 1988), paradox is proposed as a means of providing researchers, practitioners, and organisational members with a new way of thinking about, and working with, the complex nature of change. Three studies were undertaken to investigate the contribution of paradox to the study and practice of organisational change. The first study was concerned with identifying a comprehensive set of the paradoxes relevant to organisational change. This study utilised an interview-based methodology to identify change-related paradoxes for three organisational stakeholder groups, namely managers, change agents, and change participants. Ten managers, ten change agents, and ten change participants working in professional and administrative roles within a large Australian University participated in the study. Results identified 16 different change-related paradoxes, 15 of which were similar to organisational paradoxes that had been previously identified in the literature. There were both similarities and differences in paradox identification between the three stakeholder groups. Three of the 16 paradoxes were identified by all three stakeholder groups, while others were identified by two out of the three groups, or by just one group independent of the other two. Change agents demonstrated a unique ability to identify paradoxes that had also been identified by either managers, or change participants. The paradoxes uniquely identified by the different stakeholder groups seemed to be consistent with the role expectations and role identities for the groups. This study builds on existing literature by using empirical methods to specify exactly which paradoxes are relevant to the study and practice of organisational change. It also demonstrates that all three organisational stakeholder groups are able to identify change-related paradoxes. The purpose of the second research study was to better understand how each of the three stakeholder groups interpret the paradoxes relevant to organisational change. Twenty managers, 19 change agents, and 19 change participants in professional and administrative positions within a large Australian University completed a 307-item paired-comparison questionnaire suitable for multidimensional scaling analysis (MDS). The items in the questionnaire were generated using 13 of the 16 paradoxes identified in the first study. MDS was used to reveal the underlying structure of the data set for all three stakeholder groups. Results found four-dimensional solutions for all groups. The dimensions were sufficiently similar across the groups to produce a fourth solution using the three groups combined. The four paradoxical dimensions identified in this solution were: approach to change (participative versus directive); scope of change (incremental versus transformational); attitude to change (acceptance-oriented versus action-oriented), and focus of change (local-level versus system-level). The first two of these dimensions had been identified previously in the literature while the third and fourth dimensions add to the existing literature. This four-dimensional solution offers a practical framework to aid the study and practice of organisational change. External unfolding analysis was used to aid in the interpretation of the MDS solutions. This analysis considered the extent to which each of the items in the questionnaire were perceived to result in resistance to change, achieving the intended outcomes of the change, and achieving satisfying change outcomes. For all four solutions, results were significant for approach to change. Participative change was associated with more satisfying change outcomes, and directive change was associated with greater change resistance, and greater likelihood of achieving the intended outcomes of the change. The third research study was concerned with demonstrating that identification and resolution of change-related paradox is a learnable skill and can therefore be improved through training. Sixty-three participants (29 men and 34 women) occupying professional and administrative roles at a large public sector organisation participated in the study. Participants were allocated to either a control or experimental group. Case studies were used to provide time 1 and time 2 measures of both groups' ability to identify and resolve paradox. A training intervention methodology was developed based on a review of the adult learning and training literatures, as well as previously-utilised methodologies of paradox identification and resolution identified from the literature. The four-dimensional schema of change-related paradox developed in the second research study was used to define the four paradoxes in the training intervention. A 2 x 2 repeated measures MANOVA was used to test whether training improved the ability to identify paradox. Results found that training improved participants' ability to identify the approach to change and scope of change paradoxes, but not the focus of change paradox. Eight Wilcoxin Signed-Ranks tests were used to determine whether training improved ability to resolve change-related paradox. Results showed that training improved the ability to resolve the approach to change, scope of change, and attitude to change paradoxes, but not the focus of change paradox. Resolution was also found to be relatively infrequent among participants, suggesting that it is a difficult undertaking. It was concluded that a paradox-based approach to the study and practice of organisational change is useful in that it offers a new framework to inform change theory and enhance change implementation.