analyse how suspects and accused as well as other key stakeholders (and society at large) are affected by inadequate procedural rights and why the problem persists. Effective research needs to (strive to) be based on reliable and credible data and be impartial and balanced, showing different sides and perspectives of the problem. A key component of human rights research is interviews. Interviews should be held with those affected (suspected and accused persons) as well as those responsible (e.g. police officers) and those involved in the context (e.g. lawyers, doctors, judges and prosecutors). This not only improves the results of the research but also increases its credibility and the chances that the findings and recommendations are accepted by all those that need to be influenced. Moreover, interviewing a broad range of stakeholders helps to engage them from the outset (see above, § 2.4) and to build trustful relationships for the change process. Research should aim at having a societal impact. As a speaker at the EU workshop with CSOs said: “Sometimes we forget what we need our research for.” Therefore, it is useful to critically self-reflect when, and how, research is most useful and to embed research into a clear change strategy such as a Theory of Change. Thus, to achieve societal impact, we believe it is beneficial that research on the problem and its systemic causes (as conducted in the project Inside Police Custody 2)79 is ideally complemented by in-depth research on promising practices of other countries serving as models of change (see factsheets). It is further desirable that research results are spelled out in effective recommendations meeting the SMART-criteria: Specific, Measurable, Achievable, Results-oriented, and Time-Bound.80 Our project’s findings showed that research into other jurisdiction’s best practices was perceived as an extremely valuable resource by both partners and stakeholders involved. This seemed especially true when the research on promising practices was detailed enough to put stakeholders from other countries in a position to understand the main features of a practice originating from a different system. Our experience also showed that reflections and lessons learnt from the implementation of the practices were particularly appreciated, as they could be useful in replicating the practice in different national contexts. Further, we found that research on promising practices was especially effective and convincing when accompanied by testimonies of the stakeholders directly involved in the development or implementation of the practice. Another key finding of the CSOs involved in the project was the importance of close involvement with suspected and accused persons working with them rather than just on them. To this end, the methods of ‘action research’ could be particularly useful. Action research is defined as a “democratic and participative orientation to knowledge creation. It brings together action and reflection, theory and practice, in the pursuit of practical solutions to issues of pressing concern. Action research is a pragmatic co-creation of knowing with, not only about, people.”81 Designing action research starts with deciding to what extent the CSO wants to involve impacted persons. It ranges from training impacted persons to undertake the research themselves to involving participants in taking research decisions (e.g. on what to ask and how to ask it), to consult and inform them about planned actions, or to inform them about the research results.82 With the end of the project in mind, researchers do not simply investigate the present but also initiate action to improve the future.83 For this reason, action research functions with recurrent cycles consisting of reflection, planning, implementation, observation, and evaluation. These cycles are repeated until every participant of the project notices change.84 They go beyond simply presenting 79. Lloyd-Cape, ‘Inside Police Custody 2’ (2018); Zach, Katona, and Birk, ‘Inside Police Custody 2: Country Report’ (2018). 80. For a detailed description on how to draft effective recommendations see APT, ‘Briefing No 1: Making Effective Recommendations’ (2008). 81. Hilary Bradbury, ‘Introduction: How to Situate and Define Action Research’ in Hilary Bradbury (ed.), The SAGE Handbook of Action Research 3rd edn, (SAGE Publications 2015). 82. Anna Góral and others, Action Research: A Handbook for Students (Instytut Spraw Publiczynch 2021) 49. 83. Ishwara P. Bhat, ‘Action Research in Law: Role and Methods’ in Idea and Methods of Legal Research (Oxford Academic, 2020) 533. 84. ibid535. 34