Judicial psychology examines how memory, perception, communication and human decision-making affect what happens in and around courts. Its purpose is to improve the quality of evidence and fair participation—not to read someone’s mind, diagnose a litigant or replace legal judgement.
This Gold Standard guide follows eyewitness evidence, court-user experience, jury comprehension and psychological evaluation through original fictional examples. The central question is what a finding can truly establish and how its meaning changes across different legal and human settings.
THE GOLD STANDARD · JUDICIAL PSYCHOLOGY
Understand memory, judgment and the experience of fairness
Chapter index
Memory, perception and identification evidence · Chapters 1–12
- What judicial psychology can and cannot decide
- Psychology, legal rules and ethical principles have different jobs
- Begin with the decision and its human information path
- Perception is selective and cannot be replayed perfectly
- Memory is reconstructed, not a permanent video file
- Confidence changes meaning with the time it was recorded
- Questions can influence what people report
- Identification procedures are systems with measurable error
- A worked identification example distinguishes hits and errors
- Base rates influence how a result should be interpreted
- Group findings are not individual lie detectors
- Court users evaluate fairness through several dimensions
Court communication, human judgments and expert evidence · Chapters 13–25
- Voice means meaningful opportunity, not unlimited speaking time
- Respect and neutrality must be observed, not just announced
- Language access shapes what a participant can understand
- Jury instructions are a learning task
- Self-reported understanding differs from demonstrated understanding
- Attention and cognitive load can shape comprehension
- Narratives help organise information and can mislead
- Anchors and first impressions require careful interpretation
- Bias is not the same as proof of personal misconduct
- Forensic evidence needs interpretation and limitations
- Expert testimony has an explicit scope
- Emotion and stress are not simple truth scales
- Children and vulnerable participants need special care
Procedural fairness, applied statistics and transfer · Chapters 26–38
- Court interpretation has psychological and linguistic dimensions
- The physical and digital court environment affects attention
- Worked court-user survey: respondents are not all invitees
- Perceived fairness and lawful fairness are distinct outcomes
- Statistics can quantify error without judging guilt
- Decision support and AI still require human accountability
- Confidentiality and privacy affect the research design
- A court-system evaluation must use an appropriate baseline
- Second worked case: a changed instruction without a changed rule
- Diagnose a claim before turning it into reform
- Changed-context transfer: a new population or legal setting
- Frequently asked questions about judicial psychology
- Read further across the eduKate knowledge roots
1. What judicial psychology can and cannot decide
Judicial psychology examines how human cognition, memory, emotion, communication and social interaction affect the experience and assessment of information in judicial settings. It draws from psychological research relevant to witnesses, judges, lawyers, jurors, parties and court users. The field studies decision processes and human limitations; it does not provide a secret test that reveals which litigant is telling the truth or which person deserves to win a case.
Research may help evaluate how an eyewitness identification procedure influences memory, how people understand judicial instructions, or whether court interactions allow meaningful participation. But scientific findings about groups do not settle the facts of an individual case automatically. Legal admissibility, procedural duties and verdicts depend on the applicable jurisdiction and the evidence properly before its decision-makers. The gold standard states what was measured, what is inferred and what remains for lawful adjudication.
2. Psychology, legal rules and ethical principles have different jobs
Psychology describes and tests claims about human behaviour and cognition; legal rules establish the authorised processes, burdens and standards for deciding disputes; judicial ethics governs professional conduct. An empirical finding that a particular procedure reduces one kind of error does not automatically make that procedure legally compulsory everywhere. Nor does compliance with a rule guarantee that no human judgment error occurred.
Distinguish an observed cognitive effect from a claim about what a court ought to do. Psychological evidence may inform policy reform, training or expert testimony, but legal conclusions require appropriate authority and argument. The existing Judicial Studies, Judicial Ethics and Law roots provide neighbouring fields without making them interchangeable with judicial psychology.
3. Begin with the decision and its human information path
A useful inquiry asks a narrow question: which information is received, by whom, in what form, under which conditions and for which decision? A witness seeing a brief event, a jury reviewing complex documents and a self-represented litigant completing a form have different tasks. One general statement that ‘people are biased’ is not a scientific analysis of all three.
Map the sequence from event or source through perception, recording, later communication and judgment. At each transition identify possible loss, misunderstanding, outside influence or procedural barrier. An evidence map can show which details were directly observed, which were recalled later, which were reconstructed from documents and which remain uncertain. The point is to improve questions, not to prejudge people.
4. Perception is selective and cannot be replayed perfectly
Visual perception depends on attention, lighting, distance, movement and the conditions in which a person encounters an event. Human vision does not record every feature of a scene with equal resolution. A witness may accurately report a central action while missing a peripheral detail, or remember a conspicuous object while being uncertain about surrounding people.
The National Academies’ review of vision and memory explains why identification depends on sensation, perception and later memory, and why those capacities have limits. The appropriate lesson is neither to assume all eyewitnesses are unreliable nor to treat a vivid description as a flawless recording. Investigators and courts should consider the actual conditions and the legal processes governing how evidence is evaluated.
5. Memory is reconstructed, not a permanent video file
Memory can preserve useful information while also being shaped by later experiences, questions and exposure to other accounts. Recalling an event involves processes that need not reproduce every original detail. This matters particularly when someone describes a person or sequence after a substantial interval or after discussing the event with others.
The National Academies’ 2014 review documents how memory-related factors can affect eyewitness identification and recommends procedures intended to minimise contamination. Such research offers general principles, not an individual diagnosis of whether someone lied. A sincere witness can make an error; an inconsistent recollection does not by itself establish dishonesty. Preserve what was said, when it was said and which intervening information the person may have encountered.
6. Confidence changes meaning with the time it was recorded
Confidence expressed at the moment of an initial identification differs from confidence expressed months later in court. Later discussions, feedback and preparation may change what a person believes about an earlier identification. A highly confident final statement is not therefore a simple measurement of the original memory’s accuracy.
The National Academies’ eyewitness-identification recommendations recommend recording the witness’s own words about confidence at initial identification and explain why later confidence is vulnerable to outside influences. The relationship between initial confidence and accuracy is conditional on the procedure and circumstances; it cannot be reduced to a universal conversion table from certainty to truth. The gold standard preserves original conditions and avoids letting confidence replace independent evidence.
7. Questions can influence what people report
The wording, order and assumptions embedded in questions can change how respondents interpret a task. ‘Did you see the red car?’ differs from asking someone to describe vehicles they remember without supplying a colour. A leading question may communicate information not present in the original experience, particularly when repeated or combined with feedback.
Research on interviewing supports carefully designed protocols and documentation. The correct approach in real investigations is governed by trained professionals and local law, not a universal classroom script. A scientific report should explain the exact procedure used before comparing witness answers. The Critical Thinking root helps identify assumptions introduced by the question itself.
8. Identification procedures are systems with measurable error
Lineups and photo arrays are structured identification tasks, not mere tests of witness sincerity. The composition of options, instructions, administrator behaviour and handling of confidence statements can alter their evidentiary value. Design choices influence what the witness is being asked to decide and how results should be interpreted.
The National Academies’ summary recommendations include blind administration, standardised witness instructions, documenting initial confidence verbatim and recording identification processes. These are recommendations from a U.S.-centred research report, not an assertion that the same procedures are currently mandatory in every jurisdiction. The practical lesson is that evidence quality depends partly on how the evidence was elicited.
9. A worked identification example distinguishes hits and errors
Imagine a wholly fictional laboratory task with 100 image comparisons, 20 containing the predefined target under the study’s rules and 80 not containing it. Participants correctly flag 16 target-present trials but also flag 12 target-absent trials. In this exercise there are 16 true positives, 4 false negatives, 12 false positives and 68 true negatives.
The sensitivity for target-present trials is 16 ÷ 20 = 80%. The proportion of flagged trials that truly contain the target is 16 ÷ (16 + 12) ≈ 57.1%. These are different conditional quantities. The example is not an estimate of any court’s error rate, witness accuracy or probability that an accused person is guilty. A lawful real-world evaluation must consider different settings, prevalence, evidence and applicable standards.
10. Base rates influence how a result should be interpreted
A measurement may perform consistently under a particular test and still yield different proportions of correct positive flags when the mix of cases changes. This is a consequence of conditional probabilities. If the target is rare among examined items, the same false-positive behaviour may account for a larger share of all positive flags.
Do not import a psychological experiment’s prevalence into an actual court and declare a posterior probability of guilt. Cases reach courts through complex processes and contain multiple kinds of evidence. Statistical reasoning can reveal a mistaken denominator, but does not substitute for legally authorised fact-finding. The Statistics and Evidence-Based Thinking roots develop those distinctions.
11. Group findings are not individual lie detectors
Psychological research often estimates patterns across participants under defined conditions. A finding that a method increases average error does not establish that a particular individual’s report is false. Similarly, body language, eye contact, hesitation or confident speech have many possible causes and should not be presented as reliable stand-alone signs of deception.
An individual case requires its own admissible evidence and context. People differ in language, disability, culture and stress responses, making crude behavioural rules particularly risky. A science-based account is cautious about the inference it makes from an observed gesture. The strongest reasoning distinguishes a tested group relationship from an unsupported verdict about one person.
12. Court users evaluate fairness through several dimensions
Procedural fairness concerns the experience and legitimacy of the process as well as formal rules. The California Judicial Branch’s procedural fairness overview identifies respect, voice, neutrality and trust as important dimensions of how people perceive court proceedings. These are empirical and administrative concerns that sit beside, but do not replace, legal standards of due process.
A person may feel heard yet dislike a lawful outcome, or obtain a favourable outcome after a confusing experience. A survey about respect is not a direct measurement of the judgment’s legal correctness. A good court assessment asks whether people had an appropriate opportunity to participate, received understandable explanations and experienced consistent, impartial treatment.
13. Voice means meaningful opportunity, not unlimited speaking time
Voice in procedural-fairness research refers to a person’s opportunity to present relevant concerns and be taken seriously. It does not mean every court participant can determine the procedure or speak without limits. Relevant evidence rules, safety requirements and efficient case management remain important parts of an orderly process.
Psychological research can investigate whether people understand when they may speak, whether they feel listened to and which barriers limit participation. It cannot on its own prescribe the exact hearing procedure of every jurisdiction. The California Judicial Branch’s procedural fairness overview is a source on one U.S. state court system’s administrative practice; actual rights and procedures depend on the relevant court.
14. Respect and neutrality must be observed, not just announced
Polite language, consistent rules, clear directions and careful attention to affected people can support perceptions of fair process. But claiming that a court is ‘neutral’ because its website uses that word is not evidence about how interactions occur. Researchers need appropriate observation and survey methods with clear definitions.
Neutrality also has a legal meaning involving independent and impartial decision-making. A psychological survey of perceptions cannot determine every actual conflict or procedural breach. The distinction between perceived fairness and compliance with law is essential when translating human experience into policy recommendations. Refer to Judicial Ethics for normative rules about conduct.
15. Language access shapes what a participant can understand
A court user may know the facts of their dispute yet misunderstand an unfamiliar term, deadline or instruction. Technical legal language, stress, hearing difficulty, literacy and limited proficiency in the court’s language can create distinct barriers. The problem should not be attributed automatically to a lack of intelligence or concern.
Research can test whether explanations are understood in the intended setting and whether suitable language assistance improves comprehension. Professional interpretation involves accuracy, neutrality, confidentiality and the relevant legal process. A user survey alone cannot certify that a legal translation was correct. The Communication and Information Ethics roots explain how accessibility and meaning interact.
16. Jury instructions are a learning task
In jurisdictions using juries, jurors may have to understand unfamiliar legal concepts, remember evidence and apply a judge’s instructions under the authorised process. The U.S. Courts’ jury-process explanation makes clear that instructions define what jurors should and should not consider; their exact law is case- and jurisdiction-specific.
Psychological questions include how wording, organisation, written access and timing affect comprehension. A juror saying they understood a direction is not necessarily equivalent to demonstrating its correct application. Studies should define and measure both self-reported confidence and actual comprehension where suitable, without intruding into protected deliberations.
| Study classification | Target present | Target absent |
|---|---|---|
| Flagged positive | 16 | 12 |
| Not flagged | 4 | 68 |
| Totals | 20 | 80 |
17. Self-reported understanding differs from demonstrated understanding
Surveying people about whether an explanation was clear is useful, but it does not directly establish that they could apply the instruction correctly in a changed situation. Someone may feel confident about a definition while overlooking an exception, or understand a simple question but struggle when the same language is combined with other instructions. Evaluation therefore needs to specify what comprehension means.
A New South Wales juror-instruction survey involved questionnaires completed by jurors after criminal trials and explicitly recognised limitations of self-reported understanding. The result illustrates a methodological point rather than an estimate of all courts’ comprehension problems. A strong evaluation may combine user experience, appropriately designed knowledge checks and careful respect for the law governing court proceedings.
18. Attention and cognitive load can shape comprehension
Court information may arrive through lengthy spoken instructions, documents, exhibits and questions under time pressure. People have limited capacity to process unfamiliar detail at once, especially when terminology or the order of information is confusing. This suggests studying how essential distinctions are defined and when participants can refer back to relevant material.
An educational explanation can be shortened without omitting legal qualifications, but the official content of a real jury instruction or court direction must be controlled by the authorised court. Research can test whether a legally accurate restatement is clearer in a particular setting; it cannot unilaterally replace required language. The Education root develops learning and transfer without turning a courtroom into an ordinary classroom.
19. Narratives help organise information and can mislead
People use stories to connect events and infer why one thing followed another. Narrative organisation can help readers remember a complex case, but it may also tempt a listener to fill gaps or prefer a coherent account even when crucial parts lack evidence. An attractive chronology is not itself proof of cause or responsibility.
One party may present a clear explanation while another presents technical documents that are harder to follow. Evaluation needs to distinguish communication quality from evidentiary strength, without assuming all presentations should have equal persuasive effect. A psychologically informed analysis asks which link in the story is supported and where the listener may be supplying an untested assumption.
20. Anchors and first impressions require careful interpretation
An initial figure, allegation or label can affect later interpretation through mechanisms studied in judgment research. Yet the strength and scope of any anchoring effect depend on the task, context and design of the research. It would be irresponsible to assign a fixed percentage of distortion to every judge, juror or litigant.
Where appropriate, studies may test whether different sequences of the same information lead to different comprehension or valuation under carefully controlled conditions. Real decisions also have legal constraints and multiple sources of evidence. The result should identify the experimental measure rather than extrapolate a laboratory effect to an individual verdict.
21. Bias is not the same as proof of personal misconduct
Cognitive bias describes systematic tendencies in how information is processed or judged under some conditions. It does not necessarily imply dishonesty or an ethical violation by a particular person. Skilled professionals can be affected by irrelevant context, which is why the design of review processes and information access is worth examining.
The U.S. National Commission on Forensic Science human-factors note discusses when forensic scientists should consider relevant facts and when irrelevant contextual material may bias interpretation. This is a methodological distinction, not a blanket command to keep experts ignorant of everything. Researchers should evaluate which information is essential for a task and which might create unsupported influence.
22. Forensic evidence needs interpretation and limitations
Laboratory results, pattern comparisons and expert opinions can vary in methodology and evidentiary scope. A scientific result may be valid for detecting a substance yet insufficient to identify a unique person or source. The phrase ‘consistent with’ is not automatically equivalent to certainty of common origin, and an algorithmic match score is not itself a verdict.
The National Academies’ 2009 forensic science findings emphasise the need for reliable methods, transparent error information and limits on testimony. A good judicial psychology guide teaches readers to distinguish what an analyst actually measured from what a legal decision-maker must infer.
23. Expert testimony has an explicit scope
Psychological experts may help explain general research about memory, attention or comprehension when legally admissible and relevant. Their role should not be exaggerated into deciding whether a specific witness must be believed or who should win a trial. Expertise in laboratory cognition does not automatically confer expertise in every legal, cultural or clinical question.
Courts have procedures for admitting, weighing and challenging expert evidence. The National Academies’ findings on eyewitness evidence discusses scientific-framework expert testimony in the U.S. eyewitness context. A responsible educational account identifies the research population, tested phenomenon, limitations and the jurisdiction in which an expert’s opinion would be considered.
24. Emotion and stress are not simple truth scales
Stress or fear may affect attention and recall differently depending on the person, event and conditions. An emotional witness can report accurate information, and a calm witness may also make a mistake. Declaring that people who cry are truthful or people who hesitate are deceitful has no sound general basis for individual judgment.
Trauma-informed practice can matter for how people are treated and what support is appropriate, but an educational article must not diagnose witnesses or prescribe an interrogation method for sensitive situations. Qualified professionals and the governing procedures must guide such work. Psychological respect and the reliability of evidence are connected, but neither can be reduced to a facial expression.
25. Children and vulnerable participants need special care
Children and people with particular communication or support needs may face different risks of misunderstanding, suggestion or distress in unfamiliar legal processes. But variation within any group is considerable. A broad label should not be used to presume that an individual is unreliable or incapable of taking part.
Specialist assessment, appropriate communication supports and legally authorised procedures may be needed for some participants. Studies of interviewing and memory should be used cautiously when a task or population differs from the research setting. The aim is meaningful participation and sound evidence without exposing vulnerable people to unnecessary repeated questioning or pressure.
26. Court interpretation has psychological and linguistic dimensions
Interpreted proceedings carry meaning across languages through trained human communication. A participant may understand their own account accurately but struggle with an unfamiliar legal register or a poorly interpreted question. Researchers can study comprehension and interaction without equating a language barrier with poor credibility.
The relevant interpretation standard and duties depend on the legal system. Word-for-word similarity is not always the same as preserving meaning and procedural function. A psychological study should evaluate comprehension and context while recognising that legal accuracy, neutrality and confidential handling require appropriately qualified interpretation professionals. See Translation and Communication.
27. The physical and digital court environment affects attention
Signage, queuing, acoustics, seating, remote connections and access to help can all affect how court users find and understand information. A well-designed web form can improve the experience of an authorised process, while a broken audio link may prevent a participant from hearing a necessary explanation. The same technical system can affect users differently depending on language, device and disability.
Research should study the actual tasks and relevant users, not simply count the number of screens or devices installed. Perceived convenience does not establish that a process complies with law, and technical uptime does not prove meaningful participation. The Information Ethics root explains why user access, privacy and rights need explicit consideration.
28. Worked court-user survey: respondents are not all invitees
Imagine a fictional court-service survey inviting 120 users; 70 respond and 49 say they understood the next procedural step. Among respondents, the affirmative share is 49 ÷ 70 = 70%. Across everyone invited, 49 ÷ 120 ≈ 40.8% explicitly gave that answer. The remaining 50 people’s understanding is unknown, so the result cannot support the headline ‘70% of court users understand the procedure’.
Even among respondents, saying ‘I understood’ is a perception measure rather than a direct test of comprehension. Response bias, translation, case type and timing could affect what was reported. A useful next study might compare actual task completion or an appropriate comprehension check without collecting unnecessary sensitive case details.
29. Perceived fairness and lawful fairness are distinct outcomes
Respectful communication may increase trust even if a participant does not win a case. That psychological observation matters for public service, but it cannot replace judicial independence or legally required safeguards. A pleasant interaction cannot legitimise an unlawful procedure, nor does frustration alone establish that a legal rule was violated.
The California Judicial Branch procedural-fairness principles describes fairness-related dimensions from the perspective of court users. Ethical and legal assessment require separate evidence and competent authority. Reports should explicitly identify whether the outcome measured was satisfaction, comprehension, participation, consistency or compliance with a legal requirement.
30. Statistics can quantify error without judging guilt
Conditional probability, sensitivity, specificity and sampling uncertainty can help describe an experiment or evidence-assessment tool. They also create risks when readers confuse group results with an individual legal conclusion. A reported error rate for one method and population cannot automatically be applied to a different method, case selection or judicial system.
Return to the fictional 100-comparison task: 16 true positives, 12 false positives and four missed targets. The 80% sensitivity is not the chance that a person is guilty when the tool flags them. The 57.1% positive predictive value is a property of that fictional dataset and task, not the reliability of a witness in another setting. The correct denominator belongs beside every percentage.
31. Decision support and AI still require human accountability
Software may organise documents, detect patterns or generate summaries useful to court administration and legal research. But a model may omit context, hallucinate authority, misclassify a party or produce plausible language unsupported by the record. A score is not a judicial finding and must not be treated as one solely because it comes from a sophisticated system.
Responsible evaluation identifies which function is automated, which users are affected, how errors are detected and what lawful review exists. The psychological questions concern comprehension, trust, overreliance and whether users can challenge a mistake. The legal questions concern authority, procedural rights and admissibility. These must remain distinct and explicit.
32. Confidentiality and privacy affect the research design
Court files and participant surveys can contain intimate information about families, allegations, health and employment. Research interests do not automatically justify collecting all records or making them public. The appropriate access, consent, retention and ethical review depend on the institution and applicable law.
An anonymised dataset can still reveal individuals when a subgroup is tiny or records contain distinctive details. A report should use data minimisation and appropriate aggregation while preserving enough methodological information to evaluate its claims. The Ethics and Research roots provide the wider framework for legitimate inquiry.
33. A court-system evaluation must use an appropriate baseline
To examine a new information service, record what happened before the change using clear and stable definitions. A shorter queue could reflect more staff, fewer cases or changed scheduling rather than solely a new website. A high satisfaction score may be affected by who answered or which types of cases were included.
Compare relevant periods and populations, investigate changes in workloads and record the limits of causal claims. Court systems differ across jurisdiction, case type and procedural stage; a small local pilot cannot establish that one reform will improve all courts. The appropriate outcome may involve accuracy, access, timing or understanding rather than raw throughput alone.
34. Second worked case: a changed instruction without a changed rule
Imagine a fictional administrative instruction explaining when a person must return a form. Version A uses an unfamiliar phrase and a long paragraph; Version B uses a shorter sentence that preserves the authorised meaning. In a supervised simulated test, 30 of 50 participants identify the correct deadline with A and 42 of 50 with B. The observed rates are 60% and 84%, a difference of 24 percentage points.
Those figures do not prove Version B is legally accurate or will perform identically in live proceedings. A language review must establish equivalence, and the study must test whether participants were comparable and whether the task reflects actual needs. The example demonstrates how psychology can evaluate understanding while law and professional interpretation safeguard the content of the instruction.
35. Diagnose a claim before turning it into reform
Suppose a report states that a new procedure made court users more confident. Ask what confidence meant, how it was measured, who responded, what changed at the same time and whether improved confidence corresponded to better understanding. A rise in self-assured answers might be positive, but could also conceal overconfidence.
One robust approach is a claim ledger: proposed finding, supporting observation, alternative explanation, missing evidence and responsible next action. This makes it easier to repair an overstatement than a polished public narrative that hides the measurement rules. A gold-standard recommendation should follow a defensible mechanism, not only an optimistic survey label.
36. Changed-context transfer: a new population or legal setting
Imagine repeating the fictional survey in a court where many users rely on interpretation and some participate remotely. The original questionnaire and denominator may no longer describe the service well. Measures of hearing quality, translation comprehension, access and privacy become more important, while procedural requirements may differ.
Keep the empirical principles—define the task, use suitable measures, record uncertainty—but redesign the test for the new population and legally authorised process. That is genuine transfer. Reusing the first court’s 70% response figure as a prediction for another jurisdiction would not be an evidence-grounded conclusion.
37. Frequently asked questions about judicial psychology
Can psychology determine whether a witness is lying? There is no simple validated behavioural shortcut that proves deception in a specific case. Does strong later courtroom confidence guarantee a correct identification? No. Initial conditions and later influences matter. Is a perception of fairness identical to lawful fairness? No; they are related but distinct questions.
Can a statistical experiment decide a verdict? No. It can help explain error and uncertainty under its conditions but does not replace judicial authority, evidence rules or case-specific evaluation. Do all courts use juries? No; institutions vary. Can AI remove human bias automatically? No. It introduces its own design and validation obligations as well as possible overreliance.
38. Read further across the eduKate knowledge roots
Continue with Psychology, Judicial Studies, Judicial Ethics, Law, Critical Thinking and Statistics. Each contributes a different perspective on evidence, human limitations, rules and responsible decisions.
Principal sources are the National Academies’ findings on eyewitness evidence, the California Judicial Branch procedural-fairness principles and the New South Wales juror-instruction survey. All numerical cases and courtroom situations in this guide are original teaching examples, not observations about real parties, courts or legal systems. Use The Gold Standard of Education: Learning Transfers into Life to practise tracing a claim to evidence, testing its limitations and adapting the method when conditions change.
| Measure | Number | Meaning |
|---|---|---|
| People invited | 120 | Eligible invitation group |
| Survey responses | 70 | People who replied |
| Respondents reporting understanding | 49 | 70% of respondents; 40.8% of invitees |
| Unknown responses | 50 | Cannot be assumed satisfied or dissatisfied |
| Version | Correct | Participants | Observed rate |
|---|---|---|---|
| A | 30 | 50 | 60% |
| B | 42 | 50 | 84% |