Evidence-based public policy debates

The second plenary meeting of the Network was held on 5 July. At the meeting the information of the working groups was updated and the main focus of the session was the open debate on «The question of data in public policy» as an instance of gathering shared knowledge in this area. 

The closing of the meeting, with the intention of bringing together plural actors to think about security, was in charge of the De croché puppet company, who presented a play reflecting on the situation of violence in Colombia.

What’s new in the working groups

• The website continues to receive a good flow of visits and new subscribers are added every month. The blog has maintained monthly publications during the first 6 months since the creation of the network and it is encouraged to continue publishing.
• The decision of the Core group to constitute an association as the legal form of the Network was communicated.
• The Prevention and youth group explains that its work will be oriented to community prevention and to develop an exchange of experiences and good practices.
• The Police Models group has worked on the design of the data and experience collection form. They will first work by country and then systematise the global information, with the support of the repository group.

The question of data in public policy

An introductory panel was held with the participation of Daniel Hirata (Brazil), Lucía Camardón (Argentina), Diego Torrente (España) and the moderation of Cecilia Samanes.

Cecilia Samanes expressed the importance of designing security policies based on evidence, which transcends the phrases made and with a practical approach to influence concrete proposals in terms of the real prevention of violent events.

She referred to statistics as a fundamental tool for thinking about evidence-based public policies. She indicated that the «issue» of data has to do with the need to have real, quality data in order to be able to count and implement public policies.

She stressed the need for completeness of data, that there should not be a divorce between state and alternative data.

The debate focused on:

• What information do they think needs to be built?
• How should this information be constructed?
• What is the relationship between the actors you represent? (social organisation, academia, public management/administration).
• What are the strategies that they can implement, from their own spheres of belonging, towards public policy decision-makers?

Diego Torrente explained that the objective of having data (and its analysis) is to incorporate data and theory, evidence into the policy process. This requires certain prerequisites.

First, good data. For that, one has to get close to reality, to casuistry, to draw lessons from the data. Regularities are sought in order to detect the underlying causes of phenomena (at least in the form of hypotheses). The quality of the data that measures what it should, that is from regular, reliable sources, etc.

Second, skills. It is not just a matter of collecting data, but analysis is necessary. Specific expertise is required: data analysis, plan design, social networking, training, statistics etc.

Third, political support. The meaning of «data-driven policies» is twofold: data and their analysis are required, but also ideologies, partisan balances, etc., which go beyond scientific issues, are involved. If political logics are not receptive to data and its analysis, it is very difficult to carry out these projects, precisely because of this «clash of logics».

Fourth, social support. If the proposal clashes with the interests of the target groups or the general population, it will fail.

Lucía Camardón explained that the use of information for decision-making is a pending issue for the security and justice agenda, unlike other state areas (such as health, economy, etc.).

The security forces tend to be governed by «police instinct» and politicians by electoral logic to define their strategies: they do not start from a solid knowledge of the problem.

In the case of Argentina, there is great heterogeneity in terms of information, due to its federal nature. Moreover, much of the information is not publicly available. Two important experiences in standardising information are the SNIC (National Criminal Information System) and the SNEEP (National System of Statistics on the Execution of Sentences).

There are two main areas in which the information contributes to the generation of public policies. The first is a contribution to the diagnosis of certain prioritised problems. The second is to raise awareness (especially among agencies in different areas of the State) of the need to adopt the model of decision-making based on evidence. A further objective could be to question the hegemonic punitivist model.

As challenges, she identified, among others, the disarticulation and lack of coordination between state actors; the need for standardisation, socialisation of methodologies and good practices in the use of software; traceability; exchange of information, subsistence crimes such as drug dealing, lack of analysis from a gender perspective, supplying areas with significant under-reporting; exchange with NGOs and academia, etc.

And as contributions from RISE, she suggested contributing to the culture of open data; the need for a strategic view with cross-referencing of sources; the analysis of open data based on the use of software, and drawing attention to the importance of complementing with informal and qualitative data.

Daniel Hirata made three fundamental points regarding the use of data in security in Rio de Janeiro:

The first is that in Rio de Janeiro for at least 10 years there has been no public policy on security. There are no evidence-based diagnoses, no transparency on the execution of actions and no accountability to evaluate results. Therefore, instead of «policies» we should speak of security «actions».

Secondly, he pointed out that one of the major problems is the lethality of police action. Police operations are part of the security problem in Rio de Janeiro. They are very large actions by the forces: 100 to 200 men armed with rifles, armoured vehicles, helicopters used as shooting platforms, and so on. There is high police lethality, but it cannot be measured precisely because there is no public data or control, despite the fact that this is the main instrument of security action in Rio de Janeiro.

In the last 10 years, police lethality has increased by 313%, and all political channels of negotiation have been closed.

As a third element, he highlighted the importance of data generation and civil society control. The main controls were legal submissions made by civil society actors to curb police brutality. More than 45% of deaths in Rio have been caused by the police. This prompted a ruling by Supreme Court judge Fachin prohibiting police operations in Rio de Janeiro’s favelas during the pandemic (for breach of Fundamental Precept (ADPF) 635), except in «absolutely exceptional» cases with prior notice to the Public Prosecutor’s Office. And all the data work that was contributed to the case was produced by civil society.

It is alternative, unofficial data on police operations and their results. And they generated a debate in the press and society on the subject. In the last month, access to official data has been denied, political decisions are not made on the basis of evidence.

The debate gave rise to numerous interventions and reflections around different axes:

– How to promote the generation of information in highly precarious institutions (such as those operating in the penitentiary system)? Thinking that the risk matrix created in the global north does not apply to precarious institutions such as prisons.

 – The need to incorporate qualitative knowledge to complement quantitative knowledge. And the difficulties involved in convincing the political level of its importance.

– The sources of data must be varied, questioning whether the police are the fundamental actor to produce them, because the objectives of police data collection are different from the design of public policies. In this sense, the importance of citizen and territorial participation in data generation is highlighted.

– Questioning the supposed availability of data, as well as emphasising the importance of systematisation in addition to availability.

It is worth reflecting on what is meant by «evidence» and how to avoid being conditioned by a (pre)concept of evidence? It is not data for its own sake, but data oriented towards a certain strategy for action.

Data is a tool, and its interpretation and uses must be based on a rights-based approach, on social inclusion. More than the data, what is important is the diagnosis: the factors that influence the generation of the problem. This is the key for decision-makers.

O fechamento: «Marionetes sim, brinquedos de poder não».