Using Data to Remedy Bias and Improve Outcomes

Implicit biases can affect outcomes in unforeseen ways. Using rigorous data collection and analysis it is possible to identify biases and to see how they may be contributing to unwanted outcomes. (See our earlier post on this topic).

But once biases have been identified how can they be remedied? Accurate data and analysis allow your organisation to target demographic populations adversely affected by hidden inequalities. Recording and analysing data from projects (Are you collecting the right data?) means your organisation improves outcomes and ensures you are achieving the results your clients and service users need.

Improving Outcomes for Youth Offenders

Research has shown that people who enter the criminal justice system as youths are more likely to reoffend and are more likely to be in prison later in life. Likewise, studies have shown that alternatives to criminalising first time offenders, such as restorative justice approaches, can reduce the likelihood of going to prison and reduce the likelihood of reoffending.

Targeting alternatives to prosecution and custodial sentencing at demographic groups that are disproportionately likely to be imprisoned can have measurable effects on reducing bias within the criminal justice system.

Case Study: The Youth Restorative Intervention

Get the Data carried out an evaluation of the Youth Restorative Intervention (YRI), for Surrey Police and Surrey County Council’s Youth Support Service.

The YRI is a restorative justice approach to youth offending that seeks to reduce the numbers of young people entering the criminal justice system, while also ensuring that the victims of crime are satisfied with the outcome.

Instead of an inflexible process limited to a narrowly defined set of actions for officials and participants, the YRI is designed to involve all stakeholders and to be responsive to their choices. This allows the service providers to be flexible in seeking solutions, but introduces some complexities into the ways by which outcomes are measured.

A theory of change approach can be an effective way to measure outcomes in situations such as this, and it was this method that GtD took to evaluate the effectiveness of the YRI.

In its simplest form a theory of change approach to evaluating outcomes asks if there is convincing argument for why a particular approach will lead to the intended outcomes. To define the evaluation GtD asked a set of simple questions:

  1. Is it plausible? Do the evidence and common sense suggest that this initiative will lead to the desired outcomes?
  2. Is it doable? Are the economic, technical, political, institutional, and human resources available to carry out the initiative?
  3. Is it working?  Is it achieving its stated aims?
  4. Is it worth it?  If it achieved its aims, do the benefits of the project outweigh its costs?

GtD used a variety of methods to gather qualitative and quantitative data to answer these questions and provide a detailed evaluation of the YRI. They then made recommendations for future improvements to the service.

The results of the evaluation showed that a large majority of victims (91%) were satisfied with the process, that the YRI was effective in reducing the number of first-time entrants (FTEs) to the criminal justice system (189 FTEs per 100,000 young people over a one year period compared to 394 FTEs per 100,000 over the same period two years earlier), and it was successful in reducing reoffending (27% of participants in the YRI reoffended vs 33% for a control group).

The evaluation showed that the YRI was having a measurable, positive effect on the outcomes of youth offending in Surrey, and that the benefits justified the cost of the project.

What does this mean for your organisation? 

By collecting accurate data and using varied approaches to analysing and interpreting those data Get the Data can help your organisation to:

  • Identify where biases and inequalities are leading to inconsistent, negative, or unforeseen outcomes
  • Target potential projects precisely, to remedy these biases
  • Create projects with measurable outcomes
  • Improve outcomes by rigorously evaluating projects