PROBLEM

Very few people would say openly they have any kind of prejudice, but psychologists and researchers say everyone has some kind of prejudice, even without knowing. We all have what they call "implicit bias", which is commonly described as the bad feelings or expectations we have about a group of people without being conscious about them. In the implicit bias we have prejudice based in race, color, gender, age, disability or any social group.

One of the most famous researches about the theme is the Implicit Project, conducted by Harvard researchers and its objective is to study implicit social cognition (thinking and feelings out of the conscious control), using the Implicit Association Test (IAT), which is a way to identify racial prejudice. The idea behind IAT is that some concepts and categories can be more connected in our minds than others. We may more easily, and faster, associate negative words with black children's faces than with white children's.

In the last decades, racial prejudice statistics has been changing, which shows a growing tendency in fighting implicit bias and educating society to overcome differences. In England in the 80's, 50 percent of the population was against interracial marriage, this percentege has fallen to 15% in 2011. EUA experienced a similar experience: in 1958, 94% of americans were against interracial marriage, in 2013, only 11%.

In Brazil, a country known for its miscegenation, has a reduced number of interracial marriage according to 2010 Brazillian demographic census. 75% of the people that identify themselves as white live with white people, 69% of people that identify themselves as pardo live with pardo and 50% ok people that identify themselves as black marry black people. Asian males are the ones that most have interracial marriage: 38% marry asian women, 29.2% pardo women, 22% white women and 9.8 black women.

But the Implicit bias is much deeper and harder to eradicate due to human behavior complexity and social groups diversity.

SOLUTION

Based on the Project Implicit (and Implicit Association Test), we developed a multiplatform app that tests the user's unconscious bias regarding gender, ethnics and disabilities. Using Data Analytics and domain knowledge in the test results, our solution identifies group tendencies and generates insights and proposes actions on how to minimize the effects of unconscious bias for individuals and organizations. With our results, these organizations may take measures to improve their inclusion and equality actions.

HOW IT WORKS

We designed and developed a series of tests to induce users to make uncomfortable decisions seeking to unravel everyone's personal unconscious bias. The designed tests are “Crossing the Road”, “Job Interview” and “Who's Paying?”.

On “Crossing the Road”, we test users' fears and measure their reaction towards certain stereotypes. The app shows pictures of wild and dangerous animals mixed with pictures of different people, allowing us to check the accuracy and truthfulness of the results.

With “Job Interview”, users are faced with the decision of whom to pick for interviews on certain positions, such as VP of Finance, Head of Design and Secretary. For each position, the users will need to quickly pick candidates who seem fit for an interview. This test shows how age and gender bias affects entry of minorities in top positions and even in entry level jobs.

And on “Who's Paying?”, the user's choice will be deciding who's most likely to pay for a service in restaurants and retail stores. The decision will show the impact of gender bias on the decision inclusion process, but will also show how far people with disabilities are not seen as economically independent and should be treated the same.

After the tests are complete, we compile the data offering a complete analytics platform for the institutions. With this platform, institutions will learn which bias should be targeted with the help of our tailored made suggestions.

DIFFERENTIAL

The IAT is mostly based on ethnic and the users' reflexes and it is still mainly used for research. Our tests will add greatly on gender and disabilities bias by providing a whole framework for identifying and addressing different kinds of unconscious bias in institutions.

A great issue with research experiments is the lack of user focus and widespread reach. We are developing practical products with engaging experience while still generating relevant data for research. Our Analytics insights and proposed actions are also innovative and game changing.

IMPACT

Social and Cultural

People should be seen in society as equals regardless of their gender, race and disability. Everyone has the right for a job, to be socially active and to live their lives to the best, however, unconscious prejudice of some holds back people seen as “different” by the majority. 

Economic

In the last decades, the use of implicit association tests in training has become routine in big companies. Their objective is to show the team, especially the ones that have the power to hire and to promote campaigns that they may have some bias, even unintentionally. There is an obvious interest in companies to eliminate prejudice, because if the team doesn't have any prejudice and acts rationally, it is easier to hire and grow talents. Which leads to more gains and better products.

Diversity stimulates innovation. Decades of research have shown, repeatedly, that upon receiving the task to innovate, the teams with diverse members who value everyone’s contributions outperform the homogenous teams. When working in diverse groups, Columbia Professor Katherine Phillips identified that the team members work harder and more diligently, they do this to communicate and reach a consensus with other team members that might not share the same experiences and perspectives. With this, team members think deeply in order to reach better decisions. Diversity, according to Phillips, “makes us smarter”. 

NEXT STEPS 

We will create more tests seeking a broader analysis for other situations, such as asking the user to pick among a set of pictures who the boss of the group is, who is the driver of the car, among other tests.

The test results will be used for:

1) Create directed contents in order to help the population mentalize and combat unconscious bias;

2) Advise institutions creating inclusion policies.