Summary. Jury selection is the only phase of a civil trial in which a lawyer is not trying to persuade, and the most common mistake is forgetting that. Voir dire exists to identify people who cannot be fair to your client and to build a record supporting their removal, and every minute spent arguing the case is a minute not spent finding them. This guide covers how the venire is assembled and what challenges to its composition exist, the difference between a cause challenge and a peremptory strike, the Batson framework in civil cases, the mechanics of the struck jury and strike-and-replace methods, and how to design questions that surface bias. It closes with questionnaires, permissible juror research, and post-verdict issues that trace back to selection.

Here is the single most useful reframing available in this area.

You are not selecting a jury. You are deselecting one. You get three peremptory strikes in a federal civil case and an unlimited number of cause challenges you must earn. Everyone you do not remove stays. So the entire exercise is: find the people who will not listen to your client, and get rid of them — by cause if you can, by peremptory if you must, and by a record built well enough that the ones you miss cannot say later that nobody asked.

That framing rules out most of what lawyers actually do in voir dire. Explaining the burden of proof, previewing the evidence, building rapport, and delivering a soft opening statement all feel productive and none of them removes a juror who has decided that people who sue over money are the problem. The lawyers who win jury selection ask short questions, listen to long answers, and are comfortable with silence.

How the panel gets there

The source lists. Under the Jury Selection and Service Act, each federal district maintains a written plan for random selection from voter registration lists or lists of actual voters, supplemented where necessary to ensure a fair cross-section. Many districts supplement with driver's license and other records. State systems vary and increasingly use multiple sources.

The statutory policy is that litigants are entitled to juries selected at random from a fair cross-section of the community in the district, and that no citizen may be excluded on account of race, color, religion, sex, national origin, or economic status.

Challenging the array. A challenge to the composition of the venire — as distinct from a challenge to an individual juror — is governed by 28 U.S.C. § 1867, which requires a motion before voir dire begins, or within seven days of discovering the grounds, with a sworn statement of facts. The constitutional standard from Duren v. Missouri, 439 U.S. 357 (1979), requires showing that:

  1. the group alleged to be excluded is a distinctive group in the community;
  2. its representation in venires is not fair and reasonable in relation to its numbers in the community; and
  3. the underrepresentation is due to systematic exclusion in the selection process.

Taylor v. Louisiana, 419 U.S. 522 (1975), established the fair cross-section right. These challenges succeed rarely, require statistical evidence, and demand access to the district's jury selection records — which § 1867(f) provides. They are worth considering in a district whose source lists are known to be stale.

Excusals and deferrals. Judges excuse jurors for hardship before counsel is involved, and the aggregate effect on panel composition is larger than most lawyers realize. A district that liberally excuses anyone with a job produces different juries than one that does not.

Cause and peremptory challenges

Challenges for cause are unlimited and must be granted where a juror cannot be impartial.

  • Actual bias — the juror's own statements or circumstances show an inability to set aside a preconception and decide on the evidence.
  • Implied bias — a relationship or interest so likely to produce partiality that the law presumes it, regardless of the juror's assurances. A juror employed by a party, related to a party or witness, or holding a financial interest in the outcome.
  • Statutory disqualification — under § 1865, a person must be a citizen, eighteen or older, resident of the district for a year, able to read, write, and understand English adequately, free of a disabling mental or physical condition, and not under charge or convicted of a felony with civil rights unrestored.

The rehabilitation problem. A juror says "I probably couldn't be fair to a company like that." Opposing counsel or the court then asks, "But you could follow the law and decide based on the evidence, couldn't you?" and the juror, wanting to be a good citizen in a room full of authority figures, says yes.

Courts frequently accept that. But they need not, and Thompson v. Altheimer & Gray, 248 F.3d 621 (7th Cir. 2001), is the case to cite when they should not. Judge Posner reversed a refusal to strike a juror who expressed doubt about her own impartiality and then gave an equivocal assurance, observing that a judge should not accept a formulaic recantation when the juror's substantive answers reveal a bias, and that the pressure of the courtroom makes such recantations unreliable.

How to preserve a cause challenge:

  • Get the bias on the record in the juror's own words, before anyone offers rehabilitation. Ask open questions and let the answer run.
  • Do not ask the magic question yourself. "Could you set that aside?" hands the other side its rehabilitation.
  • Follow up on the equivocation: "You said you'd 'try.' What does 'try' mean here? Is this something you'd have to work at?"
  • Make the challenge, state the grounds specifically, and if denied, note the exception.
  • If forced to use a peremptory on a juror who should have been excused for cause, say so on the record and state which juror you would otherwise have struck. Preserving the harm is what makes the issue reviewable, and in federal court the erroneous denial of a cause challenge is generally reversible only if it results in a biased juror actually sitting.

Peremptory challenges. 28 U.S.C. § 1870 gives each side three peremptory challenges in a federal civil case, with the court permitted to treat multiple parties on a side as a single party or to allow additional challenges and separate the exercise. Additional challenges for alternates are governed by local practice; Rule 47(b) provides that the court may allow additional peremptories when alternates are selected.

Three is a small number. It means cause challenges do most of the work in a well-tried case, and it means the strike list must be ranked ruthlessly — not "who do I dislike" but "who is most likely to lead the room against me."

Batson in civil cases

A peremptory challenge may not be exercised on the basis of race or gender. Batson v. Kentucky, 476 U.S. 79 (1986), established the framework in criminal cases; Edmonson v. Leesville Concrete Co., 500 U.S. 614 (1991), extended it to civil litigants, holding that a private party's use of peremptory strikes constitutes state action because the litigant exercises a governmental function through the court's machinery. J.E.B. v. Alabama ex rel. T.B., 511 U.S. 127 (1994), extended it to gender. Some jurisdictions have extended the principle by rule or decision to ethnicity, religious affiliation, and sexual orientation.

The three steps:

  1. The objector makes a prima facie showing that the strike was exercised on a prohibited basis — a pattern of strikes, the questioning history, disparate questioning, or the numbers.
  2. The striking party offers a neutral explanation. The explanation need not be persuasive or plausible at this step; it must merely be facially neutral and not inherently discriminatory.
  3. The court decides whether the objector has proved purposeful discrimination, evaluating the credibility of the explanation.

What proves pretext at step three, and what to develop in the record:

  • Comparative juror analysis. A struck juror shares a characteristic with a seated juror who was not struck. This is the most powerful evidence, and Flowers v. Mississippi, 588 U.S. 284 (2019), applied it forcefully, along with attention to the striking party's history and to disparate questioning of jurors of different races.
  • Explanations unsupported by the record — "she seemed inattentive" where nothing in the transcript reflects it.
  • Reasons that shift or multiply after challenge.
  • Failure to ask the struck juror any question about the stated concern.

Practical advice for the lawyer exercising strikes. Take notes during voir dire recording the specific answer that prompts each intended strike, contemporaneously. If challenged, you will be asked to explain, and a note reading "R3 — said insurance companies deny valid claims; hesitated on damages question" is a complete answer. A memory reconstructed under pressure is not.

And for the lawyer objecting: make the objection before the venire is discharged, or the remedy becomes unavailable. Ask for a comparative analysis on the record, and identify the seated juror who shares the characteristic.

Mechanics

Who asks the questions. Fed. R. Civ. P. 47(a) provides that the court may permit the parties to examine prospective jurors or may do the examination itself, in which case it must permit the parties to supplement by further inquiry it considers proper or must submit the parties' additional questions.

This varies enormously by judge, and it is the first thing to learn about your trial judge:

  • Some conduct all questioning and allow no attorney participation.
  • Some allow a fixed period — fifteen or twenty minutes per side is common in federal court.
  • Some allow extensive attorney-conducted voir dire, particularly in state court.

Find out early. Read the standing order, ask the courtroom deputy, and ask lawyers who have tried cases before the judge. Prepare for the format you will actually get, and submit proposed questions in writing well in advance where the judge conducts the examination — a thoughtful, neutral, well-organized list is frequently adopted in substantial part.

The two selection methods:

Strike and replace (the "jury box" method). Twelve (or the panel size plus alternates) are seated in the box. Voir dire proceeds. Challenges are exercised, and each removed juror is immediately replaced by the next in line, who is then questioned. Repeat until both sides pass.

  • Consequence: you strike without knowing who replaces the juror. It rewards caution and makes the last strike genuinely risky.

Struck jury method. The entire venire is questioned first. Cause challenges are resolved. Then the parties exercise peremptories simultaneously or in alternating order from the qualified panel, and the first twelve remaining are seated.

  • Consequence: you know exactly who the alternatives are. This is a substantially better method for counsel, and it takes longer.

Ask which method the court uses, because it changes strike strategy completely.

Panel size and unanimity. Fed. R. Civ. P. 48 requires a jury of at least six and no more than twelve members, and the verdict must be unanimous and returned by a jury of at least six unless the parties stipulate otherwise. Federal civil juries typically seat eight; there are no formal alternates because all seated jurors deliberate unless excused, and the rule permits a verdict from as few as six if jurors are excused for good cause. State practice differs, with many states using twelve-person juries and permitting non-unanimous civil verdicts — often five-sixths or three-fourths.

Stipulations worth considering: a smaller jury, acceptance of a verdict from fewer than six if jurors are lost, or a non-unanimous verdict. Each trades certainty for the risk of a mistrial, and the decision depends on which side benefits from a hung jury.

Designing voir dire

The objective, restated: identify people who cannot fairly hear your case, and build a record for removing them. Secondary objectives — communicating a theme, inoculating against a weakness, establishing credibility — are real but subordinate.

Question design principles:

  • Open-ended, not leading. "What's your view of lawsuits like this one?" produces information. "You can be fair, can't you?" produces nothing.
  • Short questions, long answers. If you are talking more than the jurors, you are doing it wrong.
  • Ask about experiences, not opinions. "Have you or anyone close to you ever been sued?" is more productive than "How do you feel about lawsuits?" Experiences are concrete, memorable, and they predict attitudes.
  • Follow the thread. The most valuable answers come from the second and third follow-up, not the first question.
  • Normalize the disqualifying answer. "A lot of people feel that way, and there's nothing wrong with it — this just may not be the right case for someone who does." Jurors will not admit bias to a room that treats it as shameful.
  • Use the panel. "Who else has had that experience?" "Show of hands: who agrees with Mr. Alvarez?" This surfaces attitudes efficiently across the whole venire.
  • Do not argue the case. It wastes limited time, it draws objections, and it teaches jurors to give the answers they think you want.

Subjects to cover in a commercial or personal injury case:

  • Prior experience with lawsuits, as party, witness, or juror. Ask about the outcome and how they felt about it.
  • Prior jury service, and whether the jury reached a verdict.
  • Relationships with parties, counsel, witnesses, and the industry.
  • Attitudes toward the specific claim type — attitudes about damages awards, about corporations, about insurance, about medical care, about employment claims.
  • Ability to award money where damages are contested, including a direct question about whether a large number would trouble the juror.
  • Attitudes toward the burden of proof and, in a defense case, comfort with returning a verdict for a party who put on little evidence.
  • Sources of information and any exposure to publicity about the case or the parties.
  • Hardship, honestly explored. A juror who cannot afford to be there will not deliberate well.
  • Anything the juror has strong feelings about that they think counsel should know. This open-ended closing question is remarkably productive.

Listen for the leaders. In every panel two or three people will dominate deliberations. They speak first, others look at them, and they hold views confidently. A leader against you is worth more than two quiet jurors against you. Rank strikes accordingly.

Jury questionnaires

A written questionnaire completed by the venire before oral voir dire.

When they are worth it: long trials, publicized cases, cases involving sensitive subjects that jurors will not discuss aloud, cases with unusual subject matter, and any case where the panel is large.

How to get one. Move early, jointly if possible, and submit a proposed instrument. Judges resist questionnaires that are long, argumentative, or that duplicate oral questioning. A ten-to-fifteen-page instrument covering demographics, experience, and attitudes is achievable; a forty-page instrument with essay questions is not.

Design points:

  • Neutral wording. Every question must be one both sides can live with.
  • Mostly closed-ended for tabulation, with a few open-ended prompts.
  • A confidentiality provision — jurors answer honestly about sensitive matters only if they know the answers are not public, and the court will need to address the public's presumptive right of access, often by sealing individual responses while making blank forms available.
  • Logistics: completed before the venire arrives if possible, copied and distributed to counsel with adequate time to review, and returned to the court.
  • Use them. The value of a questionnaire is in the follow-up. Mark the responses that require oral inquiry and pursue exactly those.

Juror research

Lawyers may research prospective jurors using publicly available information. The ethical boundary is well settled.

ABA Formal Opinion 466 provides that a lawyer may review a juror's public social media presence, but may not communicate with the juror, may not send an access request (a friend or connection request), and must consider whether an automatic notification generated by the platform — telling the juror that a particular person viewed their profile — constitutes a communication. Because platforms differ and change, the safe practice is to use accounts and settings that generate no notification, or to avoid platforms where notification cannot be suppressed.

Also:

  • Do not direct anyone else — a paralegal, a vendor, a client — to do what you may not do.
  • Report juror misconduct discovered through research, which many jurisdictions treat as a duty rather than an option.
  • Check whether the court's standing order restricts research; some judges prohibit it entirely or require disclosure.
  • What is useful: prior litigation history in public dockets, professional background, public statements about the industry or the parties, and business affiliations. What is rarely useful: general lifestyle content, which invites stereotyping and predicts little.

Jury consultants and mock trials. In a case of sufficient value, community attitude surveys and mock trials identify which juror characteristics actually correlate with verdict preference in this case, which is frequently counterintuitive. A shadow jury during trial provides real-time feedback. These are expensive and they are not available in most cases; where they are, the most valuable output is usually the case framing, not the strike list.

A working sequence

Sixty days out.

  • Determine the judge's voir dire practice, time limits, and selection method.
  • Decide whether to seek a questionnaire and move if so.
  • Identify the case's vulnerabilities — the facts a juror will hold against your client before hearing anything — and design questions to surface who holds them.
  • Prepare proposed voir dire questions for submission.

Two weeks out.

  • Finalize the question outline, ranked so the most important subjects come first in case time is cut.
  • Prepare the strike criteria: what characteristics, experiences, and expressed attitudes make a juror unacceptable, written down in advance. Deciding this in the moment produces gut reactions and Batson problems.
  • Prepare a juror chart or seating grid for tracking answers.
  • Assign roles: who questions, who takes notes, who watches the panel's non-verbal reactions, who tracks the strike math.

The day.

  • Arrive early; watch the venire in the hallway.
  • Take notes with quotes, not conclusions.
  • Watch the jurors who are not speaking, and their reactions to those who are.
  • Track cause challenges and preserve every denial.
  • Confirm the strike math before exercising the last peremptory.
  • Make and respond to Batson objections on the record, with comparative analysis.

After the panel is seated.

  • Confirm the record reflects the challenges, rulings, and objections.
  • Move to strike the array or renew objections if grounds exist, before the venire is discharged.
  • Note the composition for post-trial purposes.

Selection issues that surface later

Juror nondisclosure. A juror who failed to answer a material voir dire question honestly may support a new trial. The standard from McDonough Power Equipment, Inc. v. Greenwood requires showing that the juror failed to answer honestly a material question, and that a correct response would have provided a valid basis for a challenge for cause. Both elements matter: an inaccurate answer that would not have supported a cause challenge is not enough, and honest mistakes do not qualify.

Note the drafting implication. You can only prove nondisclosure of a question that was asked. If the judge conducts voir dire and asks nothing about prior lawsuits, a juror's undisclosed history of suing employers is not nondisclosure. Submit the questions you would want to rely on later.

Juror misconduct during trial — independent research, site visits, social media posting about the case, communication with parties or witnesses. Courts now instruct specifically about internet research and social media, and the instruction should be requested if not given. On discovering misconduct, notify the court immediately rather than investigating independently.

Post-verdict juror contact. Federal Rule of Evidence 606(b) bars juror testimony about statements made during deliberations, the effect of anything on any juror's vote, and mental processes — with exceptions for extraneous prejudicial information, outside influence, and a mistake in entering the verdict on the verdict form. The Supreme Court added a narrow constitutional exception in Peña-Rodriguez v. Colorado for clear statements of racial animus in criminal cases. Local rules frequently restrict contacting jurors after trial, often requiring court permission. Check before anyone calls.

The honest limits of all of this

Two candid observations.

First, prediction is weak. Decades of research suggest that demographic characteristics predict verdict preference poorly, that attitudes predict better but still modestly, and that case facts dominate everything. A lawyer who believes they can identify the good jurors is usually pattern-matching on stereotypes, which is both inaccurate and, in the Batson context, dangerous. What voir dire reliably does is identify the extremes — the person who has already decided — and removing those is genuinely valuable.

Second, the record matters more than the intuition. The rulings that get reviewed are cause challenges and Batson objections, and both are won on the transcript. A lawyer who elicits a clear statement of bias in the juror's own words, moves promptly, states grounds specifically, and preserves the harm when a challenge is denied has done everything the appellate process permits. A lawyer with a strong instinct and a thin record has done nothing.

So: ask short questions, listen to the answers, write down the quotes, make the record, and rank the strikes by who will lead the room. Everything else in jury selection is decoration.

Sample question sets

The following are starting points, to be adapted to the case and to the time available. Each is designed to produce information rather than agreement.

Openers that surface experience.

  • "Tell us a little about what you do, and what you did before that."
  • "Have you, a family member, or a close friend ever been involved in a lawsuit — as a plaintiff, a defendant, or a witness? What happened?"
  • "Have you ever served on a jury? Civil or criminal? Did the jury reach a verdict? How was the experience?"
  • "Has anyone here ever had to make a decision at work where the facts weren't clear and you had to decide anyway? What was that like?"

Attitudes toward the claim, in a personal injury case.

  • "Some people think there are too many lawsuits in this country. Others think people should be able to go to court when they're hurt. Where do you fall, and why?"
  • "If you decided someone was injured because of another person's carelessness, would you have any hesitation about awarding money for pain and suffering? Talk me through that."
  • "Is there a dollar figure that would strike you as too much, no matter what the evidence showed?"

Attitudes in a commercial case.

  • "This case is between two businesses. Does anyone feel that businesses should work out their own disputes rather than bringing them to court?"
  • "Have you ever felt a company treated you unfairly in a contract or a purchase? What happened?"
  • "Some of the evidence will be accounting and financial records. Does anyone find that kind of material difficult or off-putting?"

Attitudes in an employment case.

  • "Have you ever supervised or fired anyone? What was the hardest part?"
  • "Have you ever left a job under circumstances you felt were unfair?"
  • "Some people believe an employer should be able to run its business as it sees fit; others believe employees need legal protection. Where do you come out?"

Burden and standard of proof.

  • "This is a civil case, so the standard is more likely than not — not beyond a reasonable doubt. If you found the evidence just barely tipped in favor of one side, could you decide for that side, or would you want more certainty?"

The closer, which is consistently the most productive question in voir dire.

  • "Is there anything — anything at all — about this case, the parties, the lawyers, or the subject matter that makes you think you might not be the right juror for this particular case? There's no wrong answer, and nobody will think less of you for saying so."

Ask it, then wait. The silence after that question is uncomfortable, and something useful usually emerges from it.

State court variations worth checking

Federal practice is comparatively uniform; state practice is not, and the differences change strategy.

  • Attorney-conducted voir dire is far more common and often far more extensive, with some states permitting hours per side and individual sequestered questioning on sensitive topics.
  • Peremptory challenges range from three to six or more per side, with additional challenges for alternates and provisions for multiple parties.
  • Panel size is twelve in many states for civil cases, with statutory or constitutional provisions permitting smaller juries by stipulation.
  • Non-unanimous verdicts are permitted in civil cases in a majority of states, commonly five-sixths or three-fourths, which materially changes both selection strategy and settlement value — a single holdout juror is worth far less to a defendant in a five-sixths state.
  • Alternate jurors are formally designated in most state systems, and the identity of the alternates may not be disclosed until deliberations begin, which affects how jurors engage.
  • Batson extensions. Several states have adopted rules going beyond the federal framework — most notably California and Washington, which have adopted procedures directing courts to disallow strikes that an objective observer could view as motivated by race or ethnicity, and enumerating presumptively invalid reasons (inattentiveness, distrust of law enforcement, dress, demeanor, and living in a high-crime neighborhood among them). In those jurisdictions the familiar "she seemed inattentive" explanation is not merely weak; it is presumptively impermissible.
  • Juror questions to witnesses are permitted in a growing number of jurisdictions, with a screening procedure. Prepare for it where allowed, because the questions reveal what the jury is actually thinking.

The practical instruction: read the state's jury statutes and the local rules before preparing voir dire, and ask a lawyer who practices in that courthouse how the assigned judge runs selection. That conversation is worth more than any treatise.

Remote and hybrid selection

Courts adopted video jury selection under necessity and a number have retained elements of it. The practice raises specific issues worth anticipating.

Access and equity. A venire questioned by video excludes people without reliable internet or a private space, which skews composition in ways that may support a fair cross-section objection. Ask how the court addresses it — many provide facilities or devices, and the answer belongs in the record.

Observation is degraded. Reading a panel depends on peripheral information: who reacts to whom, who is bored, who leans in. A grid of small video tiles removes nearly all of it. Compensate by assigning one person on the team to watch faces exclusively while another questions, and by asking more direct questions rather than relying on inference.

Distraction and confidentiality. A juror at a kitchen table may be doing three other things, and may have a family member listening. Ask about the setting expressly, and ask the court to instruct on privacy and undivided attention.

Technical failures interrupt questioning and disproportionately affect the jurors with the worst connections — often the same jurors already underrepresented. Note failures on the record.

Where courts have settled is generally on hardship screening and questionnaire administration remotely, with in-person questioning and strikes. That hybrid preserves most of the efficiency and most of the observational value, and it is worth proposing if the court has not already adopted it.

A last note on the client in the room. Jurors watch the party, not only the lawyer, from the moment the panel enters. They notice whether the corporate representative looks bored, whether the plaintiff appears to be performing, whether counsel treats the client with respect, and whether anyone at the table reacts visibly to an unfavorable answer. Prepare the client for voir dire as carefully as for testimony: sit up, look at whoever is speaking, take notes sparingly, react to nothing, and be the same person during a recess in the hallway as at the table. Several trial lawyers of long experience will tell you that this is the most reliably useful advice in this entire subject, and it costs nothing.

Related articles

This guide is provided for general informational purposes and does not constitute legal advice. Voir dire practice varies dramatically among judges and between federal and state courts, and rules on peremptory challenges, panel size, unanimity, and juror contact differ by jurisdiction. Consult qualified trial counsel and the court's standing orders before trial.