Welcome to DAATE

We are team of passionate data scientists working to provide greater transparency into the US criminal justice system

About

Our Mission

To empower legal professionals to realize fairness and equity for their clients by providing transparency into sentencing in the US criminal justice system using data science techniques.

Why DAATE?

DAATE seeks to explore: "is there is racial disparity in sentencing in the US criminal justice system?" by providing greater transparency into sentencing of Black and White Americans in the United States (US) criminal justice system through the use of data science techniques. Leveraging the Florida Department of Corrections data from 2004-2016, DAATE is intended to be an advisory tool for defense attorneys to gain additional sentencing insight about potential inequity in the Florida legal system.

By providing transparency into sentencing of Black and White Americans, a defense attorney can use DAATE to prepare for trial to seek fair and equal treatment for their client during sentencing through simple yet impactful data science driven analytics.


The DAATE MVP provides historical sentencing analytics, a view into sentencing disparity through causality, and provides sentence time predictions that a defense attorney can use to prepare for plea bargain or trial sentencing negotiation for their clients.

  • Incarcerations

    Approximately five times more Black Americans are being incarcerated than that of White Americans

  • Sentencing

    Recent research suggests that Black Americans receive as much as 19.1% longer sentences than White American offenders

Million Active Attorneys in the US in 2021

Thousand active Defense Attorneys in the US

Million Criminal Cases Filed in the US in 2020

US States Providing Data

Number of Cases Ending in Plea Bargain 94%
White Attorneys Practicing in the US 96%
Male Attorneys in the US75%
Black Americans Arrested in 2016 in the US (Note: Double their Share of the Population)27%
Black Americans in the US Prison Population (Despite Black and Latinos Comprising of 29% of US Population)57%
Black American Population In Prison for a Drug Offense (Despite rate of drug usage was comparable) 56%

Approach

View our Approach

For more details on data, architecture, models, next steps, 1) select a filter 2) hover over an item and 3) click the details icon. Or access from the dropdown menu.

  • All
  • Data
  • Architecture
  • Predictive Model
  • Causal Model
  • Next Steps

MVP Architecture (v0.01)

A highlevel overview of the DAATE MVP architecture.

Predictive Model Architecture

An overview of the DAATE MVP predictive model architecture

Data Overview

A glimpse at the Florida Department of Corrections data

Predictive Model Results

An overview of the DAATE MVP predictive model results

Data Complexity

A glimpse into the data complexity

Causal Model Architecture

A look at the causal model approach for DAATE MVP

Predictive Model Results

An overview of the DAATE MVP predictive model results

Causal Model Approach

A look at the causal model approach for DAATE MVP

Takeaway and Learnings

An overview of some of the takeaways and learnings from the DAATE MVP.

Causal Model Evaluation

A look at the causal model approach for DAATE MVP

Causal Model Results

A look at the causal model approach for DAATE MVP

Roadmap and Recommendations

An overview on the DAATE roadmap based on recommendations.

Future Architecture

A glimpse into potential DAATE architecture

Try It

View our Dashboards

Try our MVP Dashboards.

Explore Florida Department of Corrections Data

Dive into Florida Department of Corrections data from 2004-2016

Florida Sentencing Model Results

Use this dashboard to view overall causal and predictive model results for a subset of Florida circuits

Florida Sentencing Details by Judge

Use this dashboard to select a judge to see the corresponding sentencing data associated to them.

Florida Sentencing Bias & Disparity Detection Engine Results

See the results from leveraging the Bias Detection Engine on Miami data for comparison to DAATE results.

Testimonials

View our Testimonials

We are grateful for the feedback during this journey.

Michael Braga

Pulitzer Winning Journalist/Investigations Editor for the Arizona Republic

Our team of reporters and editors at the Sarasota Herald Tribune and at USA TODAY always believed that there was tremendous potential to use criminal justice data to help bring about equity in sentencing and to shine a light on judges, prosecutors and law enforcement officials who might not even understand the ways in which they are tilting justice in favor of one group or another. We are grateful that the team that built DAATE recognized that potential and created the foundation for a tool that could be used by defense attorneys to right some the injustices that have plagued the United States for hundreds of years.

Belinda Cheng

Board Member of American Equity and Justice Group

The work that the DAATE team has been able to accomplish in such a short time is remarkable. Building in statistical significance and p-values is such a great idea to help bring confidence to the disparity charts. I'm hopeful that we can leverage this great work into the Equity Dashboard that the American Equity and Justice Group has been building, and to help us gain more momentum towards bringing more equity to the legal system nationwide.

Team

Our Hardworking Team

The team that created DAATE

Jackie Nichols

Chief Architect

Robert Ling

Data Engineer

Song Park

Data Scientist

Hao Wu

Data Scientist

F.A.Q

Frequently Asked Questions

The following are the top questions that came up during our usability tests:

  • No. DAATE is not an assessment of individual judges and may not be indicative of an individual's entire judicial record.

  • We were very fortunate and grateful to all the people who assisted us during our MVP. We had a range of people from UC Berkeley instructors, journalists, and lawyers involved in the process. For a complete list please see our acknowledgement page.

  • The DAATE team has worked hard to highlight and explain the data science techniques used in the creation of the tool. We've also included a Terms Of Use page that addresses some of the concerns.

  • The DAATE team was inspired by an investigative series by the Sarasota Herald-Tribune titled 'Bias on the Bench' that found that trial judges in Florida were sentencing black defendants to more time behind bars than whites. The DAATE team decided to explore the same data but with data science techniques.

  • The purpose of the MVP was to determine the viability of the tool and if extending it to more cities would make sense. The DAATE team is in the process of evaluating the usefulness of DAATE and will provide a more detailed update on future versions soon.

  • The DAATE team had considered applying natural language processing (NLP) to court transcripts to determine sentiment analysis in the hopes that this additional information can add additional context to the sentencing of defendants.

Contact

Contact Us

If you have questions or feedback please contact us.

Email Us