Leadership research, interactive edition
Conflict Resolution Master Class
Scientific foundations and practical techniques for preventing destructive conflict, resolving disputes, and restoring working relationships
Kelly Emrick, DHSc, PhD, MBA, BSRT(ARRT)R
- 19 peer-reviewed sources
- 4 conflict types
- 8 conversation techniques
- 7 evaluation domains
- Proposed framework, not yet validated
Make disagreement workable, and protect people from harm
Conflict resolution is a leadership capability, not a single difficult conversation. It begins with an accurate diagnosis, develops through acknowledgment, listening, clear positions and serious examination of interests, and becomes durable only when decisions have owners, changes are implemented, and recurring organizational problems receive attention.
The goal is not fewer disagreements. An employee who identifies a flawed assumption, challenges an unsafe shortcut, or questions an unrealistic deadline may be providing exactly the contribution the organization needs. The leader reduces avoidable friction and destructive escalation while preserving dissent, critical inquiry, and protected reporting.
How a reasonable disagreement becomes a personal one
A staffing plan pits two legitimate concerns against each other: one leader wants predictable coverage, another wants to protect the operating budget. The meeting ends politely, and neither concern disappears.
- 1A dispute about workCoverage versus budget. Both concerns are reasonable.
- 2A decision nobody fully understandsPeople leave without the evidence, the rationale, or their part in implementation.
- 3Behavior gets reinterpretedDelays are read as resistance. Questions are read as disrespect.
- 4A judgment about peopleThe disagreement about the work becomes a verdict on the people doing it.
What prevents the drift is a working system: make differences visible, examine their causes, decide what to do, and check whether the underlying problem has changed.
Two things to diagnose, not one
State of conflict
What people disagree about
Evidence, approach, workflow, treatment of one another, standing.
Process of conflict
How they interact while working through it
Interruption, listening, acknowledgment, how criteria are set and decisions made.
A meta-analysis of 45 studies and 3,218 teams found both deserve attention when examining team outcomes (DeChurch et al., 2013). That is about 71.5 teams per study.
The evidence behind the master class
| 268 | independent samples in the updated team conflict meta-analysis Yuan et al., 2026 | |
|---|---|---|
| 24 | neonatal intensive care teams randomized to rude or neutral comments Riskin et al., 2015 | |
| 496 | three-person teams in two emotional acknowledgment experiments Bradley et al., 2026 | |
| 1,485 | participants in a preregistered field experiment on listening and persuasion Santoro et al., 2025 | |
| 0.58 | standardized mean difference for assertive communication training on speaking up Chen et al., 2023 |
The master class sequence
Ten moves, in the order a leader usually needs them. Select any move to open that section.
About the report
Master Class in Conflict Resolution: evidence-informed techniques for leaders
The report synthesizes peer-reviewed research on team conflict, conversational receptiveness, emotional acknowledgment, listening, negotiation, mediation, psychological safety, and trust repair. Contemporary evidence challenges the assumption that task conflict generally improves team performance, and experimental findings show that disrespect can impair performance under simulated clinical conditions.
Practical applications include clarifying decision authority, separating observations from interpretations, acknowledging emotion, eliciting interests, evaluating options against explicit criteria, and documenting responsibilities. The report also addresses power imbalances, recurrent misconduct, structural causes of conflict, and why agreement or fewer complaints are weak evidence of success.
Conflict resolutionLeadershipNegotiationPsychological safetyOrganizational behaviorHealthcare teamsMediation
The integrated framework is a synthesis of practices that requires prospective evaluation. Scripts, measures and the pathway are original applications of the literature and have not been validated together as a single intervention.
Name the disagreement before choosing a technique
The word conflict can conceal several different problems. A disagreement about evidence calls for a different response than a dispute over authority, and repeated humiliation calls for a different response than a misunderstanding.
The contemporary evidence also warrants caution about celebrating conflict itself. In the updated meta-analysis, all four conflict dimensions carried negative average associations with team performance, with substantial variation across contexts.
Conflict type diagnoser
Select every statement that describes what people are actually arguing about. Conflicts often mix types.
What is the argument about?
Safety gate
Select at least one statement to see the conflict type, its central question, and the initial leadership response.
Applied diagnostic aid based on distinctions in the team conflict literature. It is not a validated assessment instrument.
Table 1. Diagnosing common forms of workplace conflict
| Conflict type | Central question | Illustrative situation | Initial leadership response |
|---|---|---|---|
| Task | What should we do? | Colleagues favor different clinical or operational approaches | Compare evidence, assumptions, risks, and decision criteria |
| Process | How should work be organized? | Departments disagree about handoffs, sequencing, or workload | Clarify workflow, responsibilities, resources, and decision rights |
| Relationship | How are we treating one another? | Interactions involve resentment, personal criticism, or mistrust | Identify specific behavior and impact; establish boundaries and repair |
| Status | Whose authority or standing is recognized? | Participants compete over influence, credit, or legitimacy | Clarify authority, ensure fair participation, and address status threats |
Figure 1. Conflict dimensions and team performance
Select a row to read its estimate. Points show corrected correlations; whiskers show 95% confidence intervals.
Relationship conflict
−0.28
95% CI −0.32 to −0.24, interval width 0.08
- Independent samples
- 204
- Teams
- 17,792
- Teams per sample
- 87.2
How are we treating one another? Resentment, personal criticism and mistrust carry a clearly negative association, estimated with high precision.
Source: Yuan et al. (2026), Table 1, accepted manuscript. Correlations corrected for sampling error and unreliability. These are associations, not percentage performance losses and not effects of training.
Table 2. Estimates and uncertainty underlying Figure 1
| Conflict dimension | Corrected correlation ρ | 95% confidence interval | Independent samples | Teams |
|---|---|---|---|---|
| Task | −0.09 | −0.14 to −0.04 | 211 | 17,667 |
| Relationship | −0.28 | −0.32 to −0.24 | 204 | 17,792 |
| Process | −0.33 | −0.41 to −0.24 | 40 | 3,136 |
| Status | −0.38 | −0.50 to −0.26 | 15 | 1,264 |
Why the sample counts cannot be added
Many studies measured more than one type of conflict, so the same sample appears in several rows. The table itself proves it.
Adding the rows would count at least 202 sample entries twice and overstate the evidence base by 75.4%. The same caution applies to the 39,859 teams the rows would sum to.
Author-derived check from Table 2, confirming the report’s instruction that samples overlap and must not be added.
The strongest associations are the least precise
Confidence interval width tracks the number of samples. Status conflict has the largest average association but rests on 15 samples, and its interval is three times wider than relationship conflict’s.
Author-derived from the published interval bounds.
Successive reviews are not a trend line
2012
Approximately neutral average relationship between task conflict and performance, with contextual conditions emphasized.
de Wit et al., 2012
2026
Negative average associations for task, relationship, process and status conflict, with substantial variation.
Yuan et al., 2026
Differences between the reviews reflect changes in their evidence and methods. They do not show that conflict has become more damaging over time.
Timing matters
A 2025 meta-analysis of longitudinal studies found that earlier task conflict predicted later relationship conflict. Temporal ordering strengthens the case for early attention, though it does not establish causation. A disagreement is easier to address before people attach personal motives to it.
Yu and Wei, 2025
Prevent avoidable conflict through work design
Conflict prevention begins before anyone becomes upset. When two departments are evaluated against incompatible goals, asking their managers to communicate more respectfully may help, but the incentives keep reproducing the dispute.
Research on workplace bullying identifies role conflict, role ambiguity and high demands as relevant organizational correlates. That justifies examining the work environment, without excusing harmful conduct or claiming that one organizational change will eliminate it (Feijó et al., 2019).
Department A is rewarded for
Accommodating additional patients
Department B is held responsible for
Overtime and schedule reliability
Every extra patient that helps A’s measure strains B’s. The dispute is designed in, so the fix belongs in the design.
Prevention review
Rate how clearly your team could answer each question during ordinary work, in language employees can actually use.
Who makes the decision, and who supplies essential input?
Which standards, constraints and priorities apply?
Where do responsibilities begin and end?
How are exceptions handled?
How do unresolved concerns reach someone with authority to act?
Answered clearly
0 of 5
Rate each question to build your fix list.
A policy that names a committee but leaves no one responsible for today’s disagreement offers limited operational guidance.
Turn “be professional” into an observable standard
A vague expectation is hard to enforce consistently. Build one people can see and check.
Include these behaviors
Hard to enforce
“Be professional.”
Observable standard
“Allow the person to finish, challenge the evidence without insulting the person, and document unresolved safety concerns.”
Preserve legitimate dissent
Employees should not be required to publicly confront a powerful colleague before their concern is addressed.
- Private consultation
- Confidential advice
- Protected reporting
Why interpersonal conditions matter
A recent systematic review and meta-analysis linked hospital incivility to poorer patient safety culture and examined associations with performance and adverse outcomes. Mixed designs and substantial heterogeneity limit causal interpretation, but the findings reinforce the importance of the conditions under which information is shared.
Freedman et al., 2025
Recognize escalation early, then narrow and stabilize
Early warning signs include repeated interruption, increasingly absolute language, accusations about motives, selective retelling of events, and disagreement spreading to unrelated issues. The first intervention is to narrow the discussion; the next is to create enough emotional stability for productive work.
Stability may require slower pacing, a quieter setting, or a pause with a specific time to resume. An indefinite postponement leaves the original problem intact.
Escalation signal board
Mark the signals you are seeing. Each one points to a specific stabilizing move.
Planning a pause?
No resume time yet. A pause without one becomes an indefinite postponement.
No signals marked. Keep the discussion on the decision in front of the group.
Examine the interpretation before acting on it
Unverified attribution
“She questioned my proposal because she wants me to fail.”
A 2025 meta-analysis associated anger with more rumination, avoidance and suppression, while acceptance and reappraisal were associated with less anger. The studies span varied populations and do not establish a universal workplace intervention (Pop et al., 2025).
Reappraisal must not erase actual harm. When someone has been threatened or repeatedly humiliated, the leader’s responsibility includes addressing that behavior.
What rudeness did in a clinical simulation
In a randomized study, 24 neonatal intensive care teams were exposed to either rude or neutral comments. Participants exposed to rudeness received lower diagnostic and procedural performance ratings (Riskin et al., 2015).
- Diagnostic gap
- 0.53 points
- Procedural gap
- 0.49 points
- Participants
- 33 neutral, 39 rude
- Randomized by
- Team (24 teams of 3)
Simulated performance only. This experiment did not measure patient outcomes or any benefit of conflict resolution training.
Figure 2. Performance ratings during a healthcare simulation
The rating scale runs from 1, failed, to 5, excellent. Switch the baseline to see how it changes the visual gap.
Neutral comments Rude comments
Rude diagnostic bar as a share of neutral
75.7%
Measured above the scale floor of 1: (2.65 − 1) ÷ (3.18 − 1)
On a scale that cannot fall below 1, the gap reads larger than zero-based bars suggest. The diagnostic gap is 13.3% of the full four-point scale range; the procedural gap is 12.3%.
Author-derived. The published figure states the 1 to 5 scale beneath its zero-based axis; both views use identical means.
| Rating | Neutral mean (SD) | Rude mean (SD) | Difference | Standardized difference |
|---|---|---|---|---|
| Diagnostic | 3.18 (0.92) | 2.65 (0.69) | 0.53 | 0.66 |
| Procedural | 3.26 (0.72) | 2.77 (0.67) | 0.49 | 0.71 |
Source: Riskin et al. (2015), Tables 2 and 3; n = 33 neutral and 39 rude. Standard deviations describe variability, not confidence intervals. Standardized differences are author-derived with a pooled standard deviation at the individual level; because randomization occurred by team, treat them as descriptive only.
Acknowledge the emotion, check the meaning, and do not mistake being heard for agreeing
Across two laboratory experiments with 496 three-person teams, about 1,488 participants, verbally recognizing another member’s emotion reduced the intensity of negative emotional expression and improved conflict resolution. Acknowledgment from team leaders was particularly consequential (Bradley et al., 2026).
Acknowledgment recognizes an experience and invites correction. It does not concede that the decision is wrong, that every accusation is accurate, or that the speaker’s preferred solution must be accepted.
Acknowledgment builder
Name the emotion as it appears, tie it to the consequence, and hand the meaning back for correction.
Opening you are tempted to use
Your statement
“This change appears frustrating, especially because your team will carry the additional workload. Have I understood that correctly?”
A question keeps inquiry open until you understand the concern.
Acknowledgment does not concede that the decision is wrong, that every accusation is accurate, or that the preferred solution must be accepted.
Summarize before responding
A summary should be checked by the person whose position it represents. The closing question is not optional.
Your summary
“Your concern is that the proposed schedule assumes coverage we do not reliably have. You expect the consequence to be overtime and delayed examinations. What have I missed?”
What the listening evidence shows, and what it does not
Four experiments
High-quality listening in disagreement
Shows reduced attitude extremity, with relational connection and less defensive reflection implicated.
Does not show guaranteed workplace agreement.
Itzchakov et al., 2024
Experimental research
Perspective getting
Shows asking for another’s perspective improved prediction accuracy in some settings.
Does not show that people seek enough information on their own; barriers were identified.
Damen et al., 2021
Preregistered field experiment, 1,485 participants
Listening and persuasion
Shows better perceptions of the persuader and more processing, immediately and at five weeks.
Does not show any added persuasive effect of the narrative presented.
Santoro et al., 2025
Two laboratory experiments, 496 teams
Emotional acknowledgment
Shows less intense negative expression and better conflict resolution, especially when leaders acknowledge.
Does not show effects in longstanding conflicts, across power relationships, or over extended follow-up.
Bradley et al., 2026
Heard is not the same as agreed
A person can feel heard, understand the argument, and still disagree. Identify what the remaining disagreement concerns.
Ask which unresolved fact would change the choice, then name who will obtain it and by when. Compare evidence, assumptions and risks against explicit decision criteria.
State a clear position, show what would change it, and negotiate interests
Receptiveness should be audible. Acknowledging opposing arguments, naming genuine agreement, qualifying claims appropriately and using constructive language were associated with better partner evaluations and, experimentally, with being seen as more suitable for future collaboration (Yeomans et al., 2020).
A position states the outcome someone wants. An interest explains what that outcome is meant to protect. “We need two additional employees” is a position; “we need dependable coverage without exhausting the existing team” is an interest that may allow several solutions.
Receptive position builder
Speak firmly without pretending certainty: a position, its basis, and the evidence that would change it.
Parts present
“My current recommendation is to test the scheduling change before adding permanent positions. The available data suggest that sequencing is part of the problem. I would change that recommendation if the pilot shows persistent demand beyond staffed capacity.”
Complete: a position, its basis, and the evidence that could change it.
Position or interest?
Classify each statement. Interests open more possible solutions than positions do.
“We need two additional employees.”
“We need dependable coverage without exhausting the existing team.”
“Add five same-day slots to every MRI template.”
“Referring offices need a reliable answer when a patient is urgent.”
“Overtime has to go to zero.”
“The department has to stay within its operating budget.”
0 of 6 classified.
Five questions for an interests-based discussion
- What must each party protect?
- Which constraints are fixed, and which can change?
- What evidence should guide the choice?
- Which feasible packages address the most important interests?This question invites combinations rather than a single winner.
- What happens if no agreement is reached?
Option package evaluator
Weigh feasible options against explicit criteria. A safety gate removes any option that falls below the acceptable standard, whatever it scores.
Criterion weights, 0 to 5
| Option | Access | Workload | Cost | Reliability | Meets safety standard | Weighted score |
|---|---|---|---|---|---|---|
| Protected urgent slots | 75.0% | |||||
| Revised appointment templates | 70.0% | |||||
| Designated approver for additions | 80.0% | |||||
| Coverage changes during predictable peaks | 75.0% | |||||
| Authorized overtime | 55.0% | |||||
| Additional positions | 70.0% |
Leading package to test
Designated approver for additions, plus protected urgent slots
- Designated approver for additions 80.0%
- Protected urgent slots 75.0%
- Coverage changes during predictable peaks 75.0%
- Revised appointment templates 70.0%
- Additional positions 70.0%
- Authorized overtime 55.0%
Scores shown are illustrative starting points for the teaching case, not findings. Replace them with your team’s evidence, and keep the criteria open to challenge: a leader’s preferred financial target is not automatically an impartial standard.
Compromise requires judgment
Splitting a negotiable resource can be sensible. Splitting the difference between an acceptable safety standard and an unacceptable shortcut is not a defensible settlement.
No single tactic wins every negotiation
A major review of negotiation research emphasizes contextual complexity: cooperation and competition coexist, and power, identity and social interaction all shape the result. It cautions against assuming one tactic works everywhere.
Boothby et al., 2023
Match the technique to the situation
Eight common situations each call for a distinct move. The wording below is illustrative; adapt it to the relationship, the setting, and the seriousness of the issue.
Knowing the right sentence in a classroom does not mean someone can use it when the other person controls their assignments, evaluations or promotion. Rehearsal with feedback is what builds the behavior.
Technique finder
Choose the situation in front of you.
Technique
Describe an observable event
“The handoff occurred after the agreed deadline on three occasions. Can we examine what happened?”
Intended purpose
Establish a shared factual starting point
Table 3. Techniques for difficult conversations
| Situation | Technique | Illustrative wording | Intended purpose |
|---|---|---|---|
| The issue is vague | Describe an observable event | “The handoff occurred after the agreed deadline on three occasions. Can we examine what happened?” | Establish a shared factual starting point |
| Emotion is intensifying | Acknowledge and check | “This seems frustrating. Is the workload implication the main concern?” | Recognize emotion without assuming its meaning |
| Motives are being attributed | Separate observation from interpretation | “I know the request was declined. I do not yet know why.” | Reduce unsupported assumptions |
| Positions appear incompatible | Elicit interests | “What does your preferred option need to protect?” | Identify underlying needs and constraints |
| A participant feels misunderstood | Summarize before responding | “Let me explain your concern as I understand it, and you can correct me.” | Check understanding |
| The discussion repeats | Identify the remaining decision | “Which unresolved fact would change our choice?” | Focus further inquiry |
| Conduct becomes disrespectful | Name the behavior and boundary | “I will discuss the proposal. Personal insults need to stop.” | Protect a workable conversation |
| Agreement remains general | Specify ownership and review | “Who will do this, by when, and how will we know it worked?” | Convert intention into accountable action |
Assertive communication training and speaking up
Pooled controlled effect among nurses and nursing students, with its 95% confidence interval.
Source: Chen et al. (2023), systematic review and meta-analysis.
In plain terms
66%
chance that a randomly chosen trained participant speaks up more than a randomly chosen untrained one
Across the confidence interval that probability ranges from about 54% to 77%, so the direction is supported while the size is uncertain.
Author-derived common-language effect size, assuming normal distributions with equal variance. Sustained workplace effects and patient outcomes remain open questions.
Build the behavior, not just the vocabulary
- RehearseA realistic situation, including unequal power.
- Receive feedbackOn what was said and how it landed.
- Try againUnder a little more pressure each round.
Follow a pathway that protects people first and closes the loop
Every concern passes a safety gate before anything else. Only then does the leader clarify facts and authority, stabilize the discussion, and look for a workable resolution. When none emerges, fair decision authority or mediation takes over, and every route ends in documented actions and a review that watches for recurrence.
Walk the pathway below. Answer each decision and the route lights up, including the loop back when a problem returns.
Figure 3. Proposed pathway for addressing workplace conflict
Interactive walkthrough of the published pathway.
Step 1
Concern or disagreement
Someone raises a concern, or a disagreement surfaces in a meeting, a handoff or a complaint. Before any technique, check whether people need protection.
Route so far: Concern
- Concern or disagreement
- Immediate danger or serious misconduct? Yes: protect people and use a formal response, then review.
- No: clarify facts, interests and decision authority
- Stabilize the discussion and examine options
- Workable resolution? No: use fair decision authority or mediation.
- Document actions, owners and review
- Review outcomes and recurrence. If the problem recurs, reassess conduct, work design and resources, then clarify again.
Original conceptual synthesis from the report. It is not a validated diagnostic instrument, legal classification, or experimentally tested protocol. Formal response requirements depend on the circumstances and applicable organizational procedures.
Impasse screen: decide, mediate, or protect
Some disputes stay unresolved after careful listening and negotiation. Screen for harm first, then for fit.
Is any of this present?
What else is true?
Route
Keep working the discussion
No harm signals and no impasse conditions are marked. Return to facts, interests and options before escalating.
Explain a decision fairly
When a leader must decide, the explanation should show the options, criteria, tradeoffs, and what would reopen the question.
Decision explanation
“We considered protected urgent slots, revised templates, and two new positions. We weighed them against timely access, sustainable workload, and operating cost. We accepted slower growth in same-day volume for the next quarter. We will reconsider if urgent requests wait longer than the agreed standard for two consecutive months.”
Implementing an authorized decision is not the same as feigning agreement. Employees keep routes for raising new evidence, safety concerns, or misconduct.
Judge mediation by more than settlement
A review of workplace mediation uses these three lenses and notes limitations in the evidence base, including power imbalances that persist outside the mediation session (Munduate et al., 2022).
Repair trust through specific responsibility and changed behavior
Resolution can leave damage behind. An employee may accept a revised decision and still be concerned about how the original disagreement was handled. Repair starts with specificity: “Mistakes were made” leaves responsibility unclear, while “I interrupted you and dismissed a concern that deserved review” names the behavior.
A meta-analysis reported a positive pooled effect of apology on trust, varying with the nature of the violation and the comparison condition. It does not imply complete restoration of trust or that an apology overcomes an integrity violation (Yuan et al., 2017).
Apology builder
Connect acknowledgment of a specific behavior with corrective action.
Your apology
“I dismissed your concern before reviewing the evidence. I am sorry. I will correct the meeting record and make sure the evidence is considered before the decision is finalized.”
- Names the behavior in the first person
- Includes a corrective action
- Free of vague or conditional phrasing
The repair sequence
- 1Name the behaviorSpecific, first person, no passive voice.
- 2ApologizePlainly, without conditions.
- 3CorrectState the action and when it will happen.
- 4Change behaviorLet time and conduct rebuild trust.
- 5Repair conditionsFix the unclear authority or design that made it possible.
If a recurring conflict reflects unclear authority, an apology alone leaves employees exposed to the same problem.
Apply the framework to a radiology operations dispute
In a hypothetical outpatient imaging center, the scheduling manager wants more same-day examinations, while the MRI team reports that adding appointments routinely extends the workday. The disagreement starts as capacity and turns personal when one group calls technologists inflexible and the other calls schedulers indifferent to safe operations.
Illustrative case. It does not report an intervention at a named facility or claim a measured improvement.
Walk the case
Six moves from a personal dispute to a reviewed decision.
Move: Stabilize
Stop the personal characterizations
The leader halts the labels, “inflexible” and “indifferent,” before anything else. Neither description can be tested, and both turn a capacity question into a judgment about people.
Fact base check
No side should defend a position from recollection alone. Which facts are documented?
Documented
0 of 6
Nothing documented yet. Gather these before asking either group to defend a position.
The interests at the table
- Scheduling
- Timely access and a reliable response to referring offices
- Technologists
- A workload that reflects actual examination requirements and available coverage
- Organization
- An explicit method for making exceptions
What the review asks
- Did access improve?
- Did the change hold over time?
- Did the same dispute return?
- Could staff raise concerns without being dismissed?
Measure what leaders do after concerns are raised
A workshop can introduce useful behaviors, but sustained conflict management needs reinforcement in daily work. Interventions targeting psychological safety, speaking up and voice have been heterogeneous with mixed results, which argues for careful implementation and evaluation rather than faith in a single session (O’Donovan and McAuliffe, 2020).
Relationships between psychological safety and objective safety outcomes have also been inconsistent and context dependent (Montgomery et al., 2025). Measure what leaders do after a concern is raised, alongside perceptions of climate.
A proposed implementation sequence
- Establish a baselineBefore training begins
- Rehearse realistic situationsWith feedback and repetition
- Coach leaders during applicationOn live disagreements
- Review behavior and organizational responseAt planned intervals
Review intervals of 30, 60 and 90 days are practical planning choices, not research-derived thresholds.
Evaluation coverage builder
Mark each domain you already measure. The measures should be interpreted together.
Domains covered
0 of 7
No domains marked. Start with the measures that show what happens after a concern is raised.
Table 4. Evaluating whether conflict resolution is working
| Evaluation domain | Possible measure | Interpretation requirement |
|---|---|---|
| Timely response | Time from a concern being raised to acknowledgment and action | Faster closure should not replace adequate examination |
| Implementation | Percentage of agreed actions completed by their due date | Verify completion rather than relying only on self-report |
| Recurrence | Reappearance of the same underlying problem | Determine whether recurrence reflects conduct, resources, or work design |
| Fair participation | Confidential feedback on opportunity to be heard | Compare experiences across roles and levels of authority |
| Freedom to speak | Reported ability to raise concerns without adverse consequences | An increase in reports may reflect greater trust |
| Operational results | Relevant delays, rework, overtime, or other local outcomes | Account for concurrent changes before attributing improvement to training |
| Durable behavior | Observed use of listening, respectful challenge, and follow-through | Assess behavior beyond immediate workshop performance |
Read the signal correctly
Choose what your data show. Each pattern has more than one explanation.
- ImprovementFewer problems are occurring. Check recurrence and implementation.
- ResignationPeople stopped expecting action. Check fair participation.
- FearPeople expect adverse consequences. Check freedom to speak.
- Lack of accessReporting routes are hard to use. Check timely response.
Leader practice self-check
Ten behaviors drawn from the master class. Rate how often you use each one: 1 rarely, 5 consistently.
I separate observations from interpretations before responding.
I acknowledge emotion and check whether I understood it.
I summarize the other position and ask what I missed.
I state my position, its basis, and what would change it.
I ask what each party needs to protect.
I make evaluation criteria explicit and open to challenge.
I name disrespectful behavior and set a boundary.
I route threats, retaliation and safety concerns to formal review.
I name owners and a review date for every agreement.
I apologize for specific behavior and follow with corrective action.
0 of 10 answered
Rate all ten behaviors to see a practice index and your lowest-rated behaviors.
Unvalidated reflection aid. The index rescales the mean rating to 0 to 100, and the four bands are equal quarters of that range for pilot triage only. The report proposes no scoring thresholds.
Know which claims the evidence can carry
The report is a narrative synthesis drawn from targeted searches, verified through journal websites, PubMed records and accessible manuscripts, with priority given to systematic reviews, meta-analyses, experiments and directly relevant organizational research. It does not claim an exhaustive search, a registered protocol, or a formal certainty assessment, and no new participant data or meta-analysis were produced.
Strength of evidence by claim
Associations across settings
Team conflict meta-analyses describe associations across heterogeneous settings. They do not show that conflict causes a given outcome.
Bounded causal tests
Experiments on listening and emotional acknowledgment support causal conclusions within the tested interactions.
Simulated hazard
Healthcare simulation shows a performance hazard from rudeness, with limited evidence on long-term clinical outcomes.
Proposed, awaiting evaluation
The integrated master class sequence, scripts, measures and pathway are original applications that require direct evaluation.
A study that would test the master class
- Combine behavioral rehearsal, leader coaching and organizational follow-through, and compare it against a credible alternative.
- Assign by team where feasible, to limit contamination between people who work together.
- Measure independent behavioral assessment, implementation of agreed actions, recurrence, fairness, and freedom to raise concerns.
- Account for employees nested within teams, and separate immediate learning from sustained change.
- Measure possible harms explicitly, including suppressed dissent and pressured settlement.
- Examine which techniques work under different power relationships, cultures, professional hierarchies and levels of severity.
Verification ledger
| Check | Result |
|---|---|
| Table 2 sample rows summed against 268 independent samples | 470, so rows overlap by at least 202 |
| Riskin participants, 33 + 39, against 24 teams of three | 72 and 72, consistent |
| Bradley teams of three | 496 × 3 = 1,488 participants |
| DeChurch teams per study | 3,218 ÷ 45 = 71.5 |
| Figure 2 rude diagnostic mean relative to neutral | 83.3% from zero; 75.7% from the scale floor |
| Chen SMD as probability of superiority | 65.9%, interval 53.9% to 76.7% |
| Figure 1 values against Table 2 | All four estimates and intervals match |
All derived values are the author’s arithmetic on published figures and are labeled wherever they appear.
References
- Boothby, E. J., Cooney, G., & Schweitzer, M. E. (2023). Embracing complexity: A review of negotiation research. Annual Review of Psychology, 74, 299–332. https://doi.org/10.1146/annurev-psych-033020-014116
- Bradley, C. M., Wolf, E. B., & Greer, L. L. (2026). Getting to the heart of conflict: Emotional acknowledgment reduces the intensity of negative emotional expression and improves intrateam conflict resolution. Organization Science. Advance online publication. https://doi.org/10.1287/orsc.2024.19578
- Chen, H.-W., Wu, J.-C., Kang, Y.-N., Chiu, Y.-J., & Hu, S. H. (2023). Assertive communication training for nurses to speak up in cases of medical errors: A systematic review and meta-analysis. Nurse Education Today, 126, 105831. https://doi.org/10.1016/j.nedt.2023.105831
- Damen, D., Pollmann, M. M. H., & Grassow, T.-L. (2021). The benefits and obstacles to perspective-getting. Frontiers in Communication, 6, 611187. https://doi.org/10.3389/fcomm.2021.611187
- DeChurch, L. A., Mesmer-Magnus, J. R., & Doty, D. (2013). Moving beyond relationship and task conflict: Toward a process-state perspective. Journal of Applied Psychology, 98(4), 559–578. https://doi.org/10.1037/a0032896
- de Wit, F. R. C., Greer, L. L., & Jehn, K. A. (2012). The paradox of intragroup conflict: A meta-analysis. Journal of Applied Psychology, 97(2), 360–390. https://doi.org/10.1037/a0024844
- Feijó, F. R., Gräf, D. D., Pearce, N., & Fassa, A. G. (2019). Risk factors for workplace bullying: A systematic review. International Journal of Environmental Research and Public Health, 16(11), 1945. https://doi.org/10.3390/ijerph16111945
- Freedman, B., Li, W. W., Liang, Z., Hartin, P., & Biedermann, N. (2025). The prevalence of incivility in hospitals and the effects of incivility on patient safety culture and outcomes: A systematic review and meta-analysis. Journal of Advanced Nursing, 81(9), 5603–5622. https://doi.org/10.1111/jan.16111
- Itzchakov, G., Weinstein, N., Leary, M., Saluk, D., & Amar, M. (2024). Listening to understand: The role of high-quality listening on speakers’ attitude depolarization during disagreements. Journal of Personality and Social Psychology, 126(2), 213–239. https://doi.org/10.1037/pspa0000366
- Montgomery, A., Chalili, V., Lainidi, O., Mouratidis, C., Maliousis, I., Paitaridou, K., & Leary, A. (2025). Psychological safety and patient safety: A systematic and narrative review. PLOS ONE, 20(4), e0322215. https://doi.org/10.1371/journal.pone.0322215
- Munduate, L., Medina, F. J., & Euwema, M. C. (2022). Mediation: Understanding a constructive conflict management tool in the workplace. Journal of Work and Organizational Psychology, 38(3), 165–173. https://doi.org/10.5093/jwop2022a20
- O’Donovan, R., & McAuliffe, E. (2020). A systematic review exploring the content and outcomes of interventions to improve psychological safety, speaking up, and voice behavior. BMC Health Services Research, 20, 101. https://doi.org/10.1186/s12913-020-4931-2
- Pop, G. V., Nechita, D.-M., Miu, A. C., & Szentágotai-Tătar, A. (2025). Anger and emotion regulation strategies: A meta-analysis. Scientific Reports, 15, 6931. https://doi.org/10.1038/s41598-025-91646-0
- Riskin, A., Erez, A., Foulk, T. A., Kugelman, A., Gover, A., Shoris, I., Riskin, K. S., & Bamberger, P. A. (2015). The impact of rudeness on medical team performance: A randomized trial. Pediatrics, 136(3), 487–495. https://doi.org/10.1542/peds.2015-1385
- Santoro, E., Broockman, D. E., Kalla, J. L., & Porat, R. (2025). Listen for a change? A longitudinal field experiment on listening’s potential to enhance persuasion. Proceedings of the National Academy of Sciences, 122(8), e2421982122. https://doi.org/10.1073/pnas.2421982122
- Yeomans, M., Minson, J., Collins, H., Chen, F., & Gino, F. (2020). Conversational receptiveness: Improving engagement with opposing views. Organizational Behavior and Human Decision Processes, 160, 131–148. https://doi.org/10.1016/j.obhdp.2020.03.011
- Yu, J., & Wei, X. (2025). Whether and when team task conflict and relationship conflict transform over time: A meta-analysis based on a cross-lagged panel model. International Journal of Conflict Management, 36(2), 452–479. https://doi.org/10.1108/IJCMA-09-2024-0238
- Yuan, B., Dong, Y., & Li, W. (2017). The trust repair effect of apology: A systematic review and meta-analysis. Advances in Psychological Science, 25(7), 1103–1113. https://doi.org/10.3724/SP.J.1042.2017.01103
- Yuan, Z., Yin, J., & Sun, J. (2026). The paradox of team conflict revisited: An updated meta-analysis of the team conflict–team performance relationships. Journal of Applied Psychology, 111(2), 195–224. https://doi.org/10.1037/apl0001315