Like many of us, my colleague Erik Girvan wants to know more about whether and how implicit bias affects the mediation process. Erik and his research assistant, Lili McEntire, put together the following report.

The term “mediation” is often accompanied by words like “neutral,” “fair,” “equal,” “balanced,” and “impartial.” The American Bar Association and the American Arbitration Association devote an entire Model Standard to impartiality, defined as “freedom from favoritism, bias or prejudice.” Given that mediators are human beings, complete neutrality and impartiality are not possible. Even so, mediators strive to be as neutral and impartial as they can. We believe that this commitment to impartiality makes the process fairer and more just for participants.

What about those mediations that involve historically marginalized groups, such as people of color? Is mediation a better (more impartial, less biased, more just) alternative to litigation for these groups? In looking at published annual reports from various courthouses all over the country (see Map A, below), we were able to find 22 states that have published some numbers regarding the number of cases that went to mediation, and the number of cases that settled in the court system that year.

This is a strong start in terms of understanding the potential for settlement that mediation can provide, but it does not give us very much information that may point out strengths or weaknesses in the process of mediation when it comes to marginalized groups of people. So we also looked for demographic information. Map B (below) depicts results from states that collect demographic information from mediation participants. Our research indicates that only three of the 22 states that reported settlement rates collected demographic information from those who participated in the court mediations. Moreover, none of the three separated out the demographic information by the number of cases that did or did not reach an agreement.

Clearly we need more data. (If you have more data relevant to either of these maps or the overarching question, please send!) If mediation is supposed to provide an impartial and safe space for dispute resolution, then mediators must manage power imbalances around socio-economic status, race, gender, and education. How successful are mediators and the mediation process in this regard? After all, even mediators with a genuine commitment to neutrality and non-discrimination may be affected by implicit bias, the automatic stereotyping that we all unconsciously do.

Right now, in the absence of specific data, we are left hoping that mediators have been able to overcome their implicit biases and thus support the efforts of both parties, fairly and equitably, to express themselves and be heard. But does this happen? Our implicit biases are easily forgotten or unacknowledged, if only because they rarely match up with what we actually believe about other people. We need not consciously endorse racist beliefs to be influenced by implicit bias in situations involving people of different races, for example. This is why it is so important to track demographic information, so that we can get a better sense of (at least in the aggregate) whether and how bias may be affecting mediated outcomes.

Measuring settlement rates can be helpful in understanding the benefits of mediation, but without other information we have an incomplete picture. We need to make greater efforts to understand where our own vulnerable points may be in adhering to the Model Standards of Conduct, and attempt to consciously make choices that will help us overcome them. Only then will we actually know if mediation can be neutral, fair, equal, balanced, and impartial.

Jennifer Reynolds is an Assistant Professor at the University of Oregon Law and the Faculty Director of the ADR Center. Teaching civil procedure, conflicts of law, negotiation, and mediation, her research interests include dispute systems design, problem-solving in multiparty scenarios, judicial attitudes toward ADR, and cultural influences and implications of alternative processes. She is also a contributor to ADR Prof Blog.