Disparity and Discrimination in Criminal Justice (UGC-NET Unit II)

A gap between groups in stop rates, arrest rates, or sentence lengths is a disparity by definition, and nothing more. It is the starting observation, not the conclusion. Disproportionality is a related but distinct measure: a group's share of an outcome, such as arrests, compared against that group's share of the general population. Researchers in juvenile justice use the specific term disproportionate minority contact for this measure when applied to youth.

Neither of those is, by itself, evidence of discrimination, the narrower claim that a decision-maker treated a person differently because of race, class, gender, or another protected characteristic, holding legally relevant facts such as offense severity and prior record constant. Establishing discrimination means isolating group membership from every legitimate factor a police officer, prosecutor, or judge was entitled to weigh, which is a hard empirical task because administrative data rarely records every legitimate factor actually used.

One method criminologists use to separate these claims is stage decomposition: comparing the disparity visible at one stage of the process against the disparity already carried in from the stage before it.

A gap that first appears at sentencing, after controlling for the offense and criminal history a person was arrested and charged with, points toward sentencing-stage bias far more directly than a gap that was already present before sentencing began. NET distractors often collapse disparity and discrimination into the same claim; the syllabus tests whether a candidate keeps them separate.

Because Unit II frames criminal justice as a structure of linked stages, the same disparity question can be asked at each stage: policing, bail, charging, sentencing, and parole.

Studies of aggressive stop-and-search programs have found some groups stopped at rates far exceeding their share of the population, while the rate at which a stop of any group actually turned up a weapon or contraband did not differ enough to explain the gap in stop rates alone. That is a disparity documented at the policing stage specifically.

Bail, charging, sentencing, and parole each carry their own documented gaps, and a gap at a later stage is not automatically a new instance of discrimination; it may simply be inherited from an earlier stage.

This is why the structure of the criminal justice system, the same structure Unit II asks candidates to know institution by institution, is also the frame criminologists use to locate exactly where in the process a disparity is generated, not merely that it exists somewhere in the process overall.

Cumulative disadvantage describes how small, individually modest gaps at each stage compound rather than simply add. A slightly higher chance of being stopped, then a slightly higher chance of being denied bail, then a slightly higher chance of a harsher charge, can multiply through the sequence into a large gap in final outcomes such as time served, even if no single stage looks alarming on its own.

This matters directly for reform design, which is a favourite NET angle on this bullet: a reform aimed only at the final stage, such as sentencing guidelines, can only close the portion of the compounded gap generated at that stage. If most of the gap originated earlier, at policing or bail, a sentencing-only fix leaves the larger share of the disparity untouched.

Coordination across agencies, the second half of this unit's scope, is one reason reforms are increasingly designed to act at more than one stage at once rather than at sentencing alone.

Criminologists use four main frameworks to explain a documented gap, and NET questions often present a short scenario and ask which framework it illustrates. Differential offending holds that a recorded disparity largely tracks a real difference in how often groups commit the offenses drawing the heaviest enforcement. Differential enforcement holds instead that the disparity arises from how enforcement is deployed, such as patrol intensity or stop thresholds, independent of any true difference in underlying conduct.

Implicit bias explains a gap as the product of automatic, often unconscious associations that shape a decision-maker's judgment without their intending or noticing it, which is distinct from deliberate discrimination. Structural or legitimacy-based accounts locate the cause further back, in unequal social and economic conditions or in a group's historically strained relationship with authorities, and link that history to whether the group experiencing disparity trusts and cooperates with the system going forward.

A single documented gap can plausibly be explained by more than one of these frameworks at once, which is exactly why NET tests the ability to distinguish them rather than memorise one correct cause.