This research was explicitly designed to generate practical solutions and suggestions for future reform grounded in the experiences of practitioners so that any recommendations would be possible to implement within the realities of politics and government, with a carefully chosen approach and data collection/analysis methods.
Approach
The research followed a pragmatist research philosophy, and an appreciative inquiry approach that asked practitioners what worked, aligning with a positive public administration approach.
Data collection
Practitioners from four countries - Australia, Canada, the UK and New Zealand - were interviewed in 2022-3. This included 71 advisers for Prime Ministers Scott Morrison,, Justin Trudeau, Boris Johnson, Jacinda Ardern as well as 3 more who worked for the incoming Australian PM Anthony Albanese. 83% of the advisors interviewed had also worked for other politicians such as ministers or MPs or party headquarters before or after their PMO role and the data therefore reflects a breadth of experiences across different types and levels of offices. Interviews were also conducted with practitioners who worked in close association to them or had played a key role in innovative reforms, such as in the House of Commons member services team, Parliamentary Service and Director of a new Office of Staff Support. The questions were principally designed to identify specific positive practices that managers of advisors could use, as well as imagine a better future:
Interview questions
- What did/does your role involve in terms of goals, tasks, skills, who you worked with/for?
- What do you think is an effective way to recruit and select staffers to meet the demands of the job?
- What do you think is the best way to orientate and train political staffers?
- What do you think helps political staffers do their job effectively?
- How should human resource management for political staffers be improved in the future?
Data analysis
Data was rigorously analysed through NVivo in a series of stages that maximised the grounding of findings in the data and interpretive validity. Stage 1 recorded initial themes in memos during interviews which were used to create an initial code book in NVivo. In Stage 2, initial open and dynamic coding of data was conducted, with each transcript coded line by line, adding new codes and changing hierarchy as the data demanded. In Stage 3, the material within each code was reviewed, and material moved to more appropriate codes where needed, codes with low numbers of references were merged and those with a high number broken up into more specific codes.