Re-designing the Integrity Management Framework in the British, Chinese and Hong Kong Public Services
Professor Paul Heywood
This study provides a comparative analysis of how the national framework for managing integrity in the UK, China and Hong Kong has been re-designed; how attitudes towards integrity have been conceptualised; and whether such moves have been helpful or harmful towards ensuring the highest possible standards of integrity.
The project used a process-tracing methodology, supported by a small number of qualitative interviews with senior officials and a quantitative survey of high-ranking civil servants.
The issues set out below deal primarily with the UK dimension of the research.
As this study was in part concerned with unethical - and potentially criminal - behaviour in public life, the research has presented a number of ethical challenges. Given that the focus of the research was senior level public officials, there was a clear possibility that any unethical behaviour they engaged in would be directly harmful to citizens as a whole. There was therefore a fine balance to be struck between our obligations – as researchers – to the research participants, and our obligations – as ethical citizens – to the population more widely.
It would be problematic to identify potentially corrupt individuals or their (alleged) crimes, and it was deemed unethical by the research team to release specific allegations against named individuals. We therefore decided to offer our research participants complete anonymity, but still proceeded to use the information gathered in a way that could identify corrupt practices. This ensured that no individual participant would be harmed by their involvement in our research, whilst allowing us to fulfil our obligations to those who are the victims of illegal acts.
The participant introduction to the survey explicitly detailed the nature and aims of the research and the potential beneficiaries. Given that the target group of respondents was high-ranking knowledgeable civil servants, participating in the research after introduction to the terms was taken as informed consent.
The key ethical issue we faced in relation to the quantitative survey was anonymity. To recruit respondents who might themselves be part of a corrupt network, whilst maintaining the principles of informed consent, it was necessary to be able to provide some guarantees about data security and anonymity, especially when dealing with senior members of the public service.
Anonymity can be particularly difficult to ensure; because of small population sizes it becomes relatively easy for motivated persons to de-anonymise data. Moreover, because the survey was conducted online, civil servants would most likely access the survey through the Government Secure Intranet. In this case, a log of the IP address of participants could potentially allow for de-anonymisation.
We took a strong line with anonymity; we grouped together Senior Civil Servants at Grades 1 – 5, creating a large enough pool to ensure participants could not be easily identified from demographic data and details of their specific department were not included. We were also careful about the metadata we collected, being mindful of the potential for inadvertently causing someone to be identified.
Researching human participants can often be a difficult task ethically; this is especially so in situations in which research participants are in a position of power, where ensuring anonymity is particularly challenging and there are allegations of wrongdoing.
In such circumstances, researchers need to consider their position both as ethical researchers and as ethical citizens: questions such as ‘to whom do we owe obligations?’ and ‘what are our ethical obligations?’ have to be paramount. We would therefore advise researchers to:
- fully consider their obligations to both the public and their participants
- develop a methodology that can offer plausible guarantees to participants, even against a motivated intruder
- work with public organisations and national governments in order to gain access, but do not lose control of the data – either in terms of ownership, or in terms of the data (and metadata) collection procedures.