Discipline Data in Florida Prisons — What Gets Punished, How, and Who?

The data behind Florida’s prison discipline system is both sprawling and opaque. Since 2018, the Florida Department of Law Enforcement (FDLE), through its Criminal Justice Data Transparency (CJDT) initiative, has published disciplinary records for adult state prisons. Yet these records have never been systematically analyzed or made accessible to the public -- not by the state, not by watchdogs, not by academics.

This project takes a ground-up approach: examining the raw data from FDLE and the Florida Department of Corrections (FDOC) to map out what gets punished, which disciplinary actions are imposed, and who is most affected. It is one of the first public investigations to offer an open, reproducible look at discipline inside Florida’s prisons -- not in summary or policy, but in the record itself.

Why This Matters

There is no official public reporting on disciplinary actions in Florida’s adult prisons. Even nationally, almost no state or federal agency publishes detailed breakdowns of prison discipline by infraction, demographic group, or outcome.

The Prison Policy Initiative’s 2025 report, Bad Behavior: How prison disciplinary policies manufacture misconduct, is essential reading for anyone trying to understand how prison discipline works (or doesn’t) in the U.S. Their work maps out the national landscape, untangles the policy logic, and shows why this data matters -- in ways that no agency ever has. If you care about the bigger picture, read their report first.

What this project adds is simple but rare: a record-level look at Florida’s prison discipline data, as held by the FDLE. It doesn’t try to reinterpret or outdo PPI’s analysis; instead, it fills in the Florida-sized blank spot in the national map by making the raw events and their disparities visible, for anyone to see, use, or challenge.

This project represents a rare -- and possibly first -- public analysis of Florida’s prison discipline system at the record level. In that sense, it both breaks new ground for Florida and adds a data-driven, local layer to what PPI and others have called a “black box within a black box.”

Why does this matter? Because understanding who gets punished for what, and how, is vital for anyone concerned about fairness, bias, and the actual day-to-day operations of incarceration. As journalists and researchers at PPI and scholars like Katherine Becker have shown, the systems governing discipline inside prisons are often opaque, highly discretionary, and riddled with racial, gender, and procedural inequities.

Official dashboards typically report only population totals and sentencing, leaving the lived reality of prison discipline unexamined. National survey data show that over half of all people in state prison are written up for a rule violation each year, and almost all receive a sanction. Yet the details -- what rules, what punishments, and for whom -- are almost always hidden from public view.

This project is a first step in pulling back the curtain in Florida, making the workings of prison discipline transparent, and connecting local reality to the national picture.

About the Data: Caveats & Ambiguities

While the FDLE’s CJDT dataset offers the most granular look yet at Florida prison discipline, it comes with serious limitations:

• Multiple Files & Overlap: The raw disciplinary data came in two files, each covering different (and sometimes overlapping) time spans. These files contain some duplicate records; see Methods for details on how overlaps and duplicates were resolved.
• Date Ambiguity: The main date field (INCAR_SUPERV_CREATED_DATE) is not when the infraction happened, but instead appears to be when FDOC sent the record to FDLE. That means we can’t actually analyze discipline timelines or trends.
• Facility Ambiguity: The “facility” field (CURRENT_INSTITUTION_ORI_TYPE_DESC) only tells us where someone was when the data was sent, not where the incident occurred. Frequent transfers make meaningful facility-level analysis impossible.
• Coverage Uncertainty: We don’t know if these records include everyone who was incarcerated during the covered periods, or just those still in custody when the data was pulled. People released or transferred might be missing.
• No Detail on Outcome Severity: “Disciplinary confinement” could mean 6 days or 60 days -- there’s no way to tell from these records. The severity and duration of punishment are not tracked here, only the type.
• One Outcome per Infraction (Not likely): The data is structured so that each infraction lists a single outcome (e.g., confinement). But in practice, almost every infraction results in multiple punishments -- for example, a person might receive disciplinary confinement, loss of gain time, and future gain time loss all for a single incident. Based on my experience, I believe the dataset records only the most severe outcome for each infraction. That means the actual number of disciplinary sanctions is much higher than reported here.
• Most Confinement Is Administrative, Not “Disciplinary”: In my own experience, only a fraction of “confinement” time is ever due to an actual infraction that goes through due process and results in a disciplinary record. Most time spent in confinement is administrative: pre-hearing, under investigation, or for “security” reasons. For example, I spent 120–150 days in confinement during my sentence, but only 30 days were for "disciplinary" sanctions that would have shown up in this data. The rest were “investigative” holds, including a single stretch of 77 days in confinement without a disciplinary charge ever being levied against me.
• Closed Management & Mass Isolation Not Captured: While “disciplinary confinement” appears very often as an outcome in these records, the data does not include the thousands of people held in “Closed Management” (CM) or “extended administrative confinement” -- Florida’s version of long-term isolation, where entire units or institutions hold people for months or years without any new infraction (Fla. Admin. Code Ann. R. 33-601.800 - Close Management). Most people in CM are isolated based on prior behavior, allegations, or simply being labeled “troublesome,” with no new disciplinary case. This means Florida’s real use of solitary/lockdown is much higher than what’s visible here.
  

All grouping, aggregation, and analysis here is based on these imperfect fields. Where fields were ambiguous or inconsistent, I’ve prioritized transparency and reproducibility. For full details, see the Methods page (Jupyter notebook).

Rule Violations

The table below lists the rule violations recorded in Florida prison disciplinary data. Scroll or search to explore more.

The top three violations -- “Disobeying verbal or written order” (12.7%), “Disrespect persons of authority” (6.6%), and “Disorderly conduct” (5.8%) -- account for 25% of all disciplinary events statewide.

These are not “dangerous” or “violent” infractions, but classic examples of what the Prison Policy Initiative calls “compliance” and “respect” violations: broad, highly subjective rules that can be used to punish people for almost any perceived disobedience, disrespect, or disruption.

As PPI notes: “These rules often serve as a ‘catch-all’ for anything staff deem undesirable, and give correctional officers extraordinary discretion to punish people for nonviolent, everyday behavior.”

Note: There are dozens of distinct rule violations used in Florida’s prison disciplinary system -- far more than can be easily summarized in a short table. For those interested in the official language and legal basis for each infraction, you can review the full disciplinary code: 33-601.314 Rules of Prohibited Conduct and Penalties for Infractions (PDF, via Prison Policy Initiative) .

Visualizing the Most Common Violations

This chart shows the 25 top rule violations by event count, matching the data in the table above. Hover bars for details.

Disciplinary Outcomes (Sanctions Imposed)

This table shows how often each type of disciplinary outcome (“sanction”) was imposed in the available FDLE records. Scroll or search to explore more.

The top three sanctions -- Disciplinary Confinement (52.36%), Loss of Gain Time (21.36%), and Cashless Canteen Denial (9.53%) -- together account for over 83% of all punishments imposed statewide.
Disciplinary Confinement -- that is, isolation or “lockdown” -- is by far the most common punishment reported here, despite widespread evidence of its harm.

Note: Despite the widespread use of the term “solitary confinement,” the Florida Department of Corrections does not technically operate disciplinary “solitary” cells. Nearly all disciplinary confinement in FDOC is in double-occupancy (“two-man”) cells -- which can be just as, or even more, psychologically damaging. In reality, though, people often end up in a cell alone for days or weeks

Based on both my personal experience and national reporting, there are also significant numbers of people in so-called “Closed Management” units, who may spend months or years in extended isolation -- often without any new infraction. These cases are **not counted** in the disciplinary data analyzed here, making the system’s real use of isolation far more widespread than what official records suggest.

As PPI notes: “Extreme punishments like solitary are not reserved for the most serious rule violations -- they are routine, imposed for minor, subjective infractions as well as for serious ones.”

Visualizing Sanctions by Frequency

This chart shows the most common types of disciplinary sanctions (outcomes) imposed on people in Florida state prisons. Hover bars for details.

Discipline Sanctions by Violation

The table below shows the number of times each type of disciplinary action was imposed for each violation. Use the search bar or page controls to explore all violations and sanctions. Download the full CSV here (includes all violations and disciplinary actions)

Who Gets Disciplined? Demographics and Disparities

It’s critical to understand who actually receives disciplinary infractions in Florida prisons. To investigate this, I joined FDLE demographic records onto the discipline data, combining fields for race, ethnicity, and sex.

For the purposes of this analysis:

• “Majority” = people identified as white and not Hispanic/Latino
• “Minority” = everyone else (Black, Hispanic/Latino, Asian, Native American, or unknown/other)

The table below shows the proportion (%) of people who ever received at least one disciplinary infraction, by race/ethnicity group and sex.

Put simply: Minority men are about twice as likely to have been written up for a rule infraction as white women. And both race and sex strongly shape who gets disciplined in Florida prisons, even before you break things down by specific rule or punishment.

Note: “Disciplinary infraction” here means at least one rule violation on record in the FDLE data associated with that person. See the Methods section for more on how these groups were constructed and tested for significance.

Racial Disparity in Discipline: Who Gets What Sanction?

The chart below shows, for every combination of rule violation and disciplinary outcome, whether minorities were significantly over- or under-represented compared to the overall prison population.

• Orange bars: Statistically significant underrepresentation of minorities (relative to baseline FDOC racial composition as recorded in the FDLE data).
• Purple bars: Statistically significant overrepresentation of minorities.
• Gray bars: No statistically significant difference.

Bars are sorted so that the most common actions and violations appear at back left.

From Over 100 Rule Violations to Meaningful Categories

The chart above shows just the top 25 rule violations -- and already, it’s a blur of overlapping labels and fine print. In reality, Florida’s disciplinary system tracks 104 distinct rules that people inside can be punished for. It’s overwhelming by design.

Why does this matter? Because, as Prison Policy Initiative’s “Florida’s Bad Behavior” report makes clear, prison systems maintain sprawling lists of “infractions” -- many vague, duplicative, or arbitrary -- giving staff enormous discretion to punish almost any behavior. The most common infraction by far, “Disobeying verbal or written order,” is a perfect example: over 1 in every 8 discipline cases is someone being written up simply for not 'following an order', when there should be more than enough rules to cover every potential order!

This avalanche of rule codes creates confusion and unaccountable power. It also makes analysis (and public oversight) difficult -- which is why I collapsed the full list of violations into a handful of more meaningful categories based on type and seriousness. This step makes the system’s patterns more transparent.

To make sense of this complexity, I grouped all 100+ granular rule violations into a smaller set of meaningful categories -- a process explained in detail below. The interactive 3D visualization immediately beneath this section uses these collapsed categories (both for rule violations and disciplinary outcomes) to show the patterns of punishment and disparity more clearly.

Scroll down after the visualization for a full breakdown of exactly how each rule was grouped and why. (Click each category to expand for details.)

Interactive Visualization

Violation Categories

Outcome Categories