Conflict Styles: Five Ways to Lose, and What Each Is For
PsyCove Editorial · 12 min read · Updated 2026-10-11
- Style is a choice made under conditions, not a personality. The same person can be entirely different with a boss, a partner and a stranger in the same week.
- There is no best style. High cooperativeness is not maturity, and high assertiveness is not strength — they are situational tools.
- Competing and collaborating both work when the outcome matters a lot. Avoiding and accommodating both work when it does not.
- The most expensive style is habitual collaborating on things that should be contested, and habitual avoiding on things that should be raised.
- What you do after the argument matters more than what you did during it. Repair behavior predicts relationship outcomes more consistently than conflict style does.
The two axes
The model asks two questions about any given disagreement. How much do I want my own outcome here — assertiveness? And how much do I want to preserve the other person's outcome and our relationship — cooperativeness? Both high is a hard combination, which is exactly why collaborating is the rarest and the most effortful mode. Both low is easy, which is why avoiding is so common.
What the axes do not capture is as important as what they do. They tell you nothing about what you say. You can be highly assertive and completely silent, or highly cooperative and relentlessly blunt — and a purely two-axis model would score both as the same person. Directness and repair behavior are the variables that get added when a model has to explain why two people with identical styles produce completely different conversations.
The five modes, and what each is for
Competing — high assertiveness, low cooperativeness
Competing pushes for your outcome and treats the relationship as secondary. It is the right tool when the issue is genuinely non-negotiable, when a decision has to be made and someone has to be accountable, and when speed matters more than agreement. Its characteristic failure is winning the exchange and losing the ability to raise anything with that person for the next six months.
Collaborating — high on both
Collaborating seeks an outcome that satisfies both sides, which means it requires diagnosing the actual interests underneath the stated positions. It is the right tool for problems that will recur — a colleague you will work beside for years, a shared project, a family pattern. It is also the slowest mode and the one most often skipped, because the cost falls entirely on the person doing the work while the benefit is invisible.
Compromising — moderate on both
Compromising trades concessions rather than solving anything. Both sides give something up so the argument can end quickly. It is the right tool for one-time decisions under time pressure, for genuinely low-stakes disagreements, and for cases where the time required to collaborate would exceed the value of the outcome. Its risk is that the trade is unequal and the person who concedes more decides so silently.
Avoiding — low assertiveness, low cooperativeness
Avoiding sidesteps the issue. It is the right tool when the issue is trivial, when the timing is terrible, when emotions are too hot to think in, and when another person genuinely needs to own the decision. It becomes costly when avoidance is habitual: unresolved small friction accumulates into the specific kind of resentment that makes a relationship feel cold without anyone identifying why.
Accommodating — low assertiveness, high cooperativeness
Accommodating yields on your own outcome to preserve the relationship. It is the right tool when you are genuinely wrong, when the issue matters more to them than to you, and when you are holding a resource they need more urgently. It becomes costly when you do it repeatedly without saying so — accommodating without naming it converts a disagreement into a record of quiet resentment.
The question is never which style is best. It is which style this particular disagreement is paying you for.
The two process variables that change the picture
Directness is about whether the thing gets said. High directness means naming the issue plainly, in words, at the time. Low directness means it surfaces indirectly — through tone, through lateness, through what is not mentioned, or not at all until it has been brewing long enough to distort everything else. Low directness is the more costly of the two, because an unspoken issue does not stay unspoken; it leaks, and the other person reads the leak and concludes you are angry about something else.
Repair is about what happens afterwards, and it carries more weight than anything measured during the argument. Repair means returning to the disagreement later, acknowledging it, and putting it to rest. Low repair does not mean the relationship ended — it means the argument is stored. Stored arguments are reactivated by the next unrelated incident, which is why a relationship can break over something trivial years after a serious fight that was never closed.
- High directness plus high repair is the combination worth building, whatever your preferred mode is.
- Low directness is the pattern most worth changing first, because it corrupts the information your partner is working from.
- High repair can compensate for a lot: a poor argument that gets properly closed damages less than a good argument left open.
- Repair does not require agreement. Coming back to say 'we didn't finish this, and I'd like to' is repair, even if nothing changes.
What actually predicts good conflict outcomes
The research is not kind to the idea that some styles are better. De Dreu and Weingart's meta-analysis found that what predicts team performance is the content of the conflict, not the style: conflict about the task tends to help teams, conflict about relationships tends to hurt them, and the effect sizes here are among the more reliable findings in this area. In other words, arguing about the work helps. Arguing about each other does not. Whether you do that while collaborating or while competing is secondary to which one you picked.
For relationships specifically, the finding that survives replication is about repair. Couples who can interrupt a conflict, take a break, and resume without the resumption becoming an accusation do substantially better than couples who do not — and the mechanism is not warmth, it is predictability. Knowing that the argument will end properly makes it safe to raise things.
Why this test is an adaptation, not the original instrument
The original Thomas-Kilmann Conflict Mode Instrument is a self-report with a scoring key that sorts respondents into one of five profiles. This site adds two dimensions to it: directness and repair. Both are process variables widely discussed in conflict research, but neither is part of the original TKI, so results should be read as a TKI core with process refinement. That is a deliberate choice — the two original axes alone cannot distinguish, for example, silent avoidance from loud avoidance, and those two need different things from you.
The other honest limitation is that this measures self-reported behavior in general, which is subject to the usual social-desirability effects. Reverse-worded items are included to dampen the tendency to describe yourself as composed. And as with any self-report, the gap between what you report and what you do is where the most useful information lives: if you read as high on cooperation but low on repair, look at the last argument you had rather than at the number.
What conflict style is not
It is not an intelligence measure, not a personality trait, and not a maturity ranking. Competing is not aggressive and accommodating is not weak — both are choices, and the choice is sometimes the right one. Nor is any of this a diagnosis, a relationship outcome, or grounds for a hiring decision.
The most common misreading is treating your style as fixed. It is far more situational than that: authority, stakes, relationship closeness, and time pressure all shift it.
Anyone who describes their conflict style with a single adjective has usually only sampled one context.
Using a result
- Identify the mismatch, not the style. Ask where your habitual mode is costing you something specific, rather than which style you should adopt.
- The lowest-value change is usually adding directness: name the thing once, plainly, at the time.
- If your repair is low, build the simplest possible version — a standing agreement that any argument can be paused and resumed within a day.
- Watch the pattern across relationships, not one. A style that works at work and fails at home is usually the same style with different stakes.
- Do not use this on anyone else without telling them. Being managed by someone's conflict style is a thing people notice and resent.
Frequently asked questions
Which conflict style is the healthiest?
None of them is healthiest. The Thomas-Kilmann framework treats the five modes as situational tools: collaborating suits recurring problems, competing suits genuinely non-negotiable decisions, compromising suits low-stakes time-pressured ones, avoiding suits bad timing, and accommodating suits situations where you are wrong or the issue matters more to the other person. What is unhealthy is using one style for every situation.
Is avoiding conflict bad?
Not inherently. Avoiding is the right move on trivial issues, at bad timings, and when someone else needs to own the decision. It becomes costly when it is habitual, because unresolved friction accumulates into resentment. The problem is the habit, not the mode.
Is competing or accommodating more mature?
Neither is more mature — that framing is the most common error in reading these results. Both are choices that are correct in some situations and costly in others. What matters is whether you chose the mode deliberately or drifted into it.
Why does this test measure more than five styles?
The original Thomas-Kilmann instrument sorts people into five profiles from two axes. This test adds directness and repair as process variables, so it produces sixteen profiles instead of five. That is a deliberate adaptation: the two original axes cannot distinguish silent avoidance from loud avoidance, and those two require different responses.
Does conflict style stay the same across relationships?
No. Authority, stakes, closeness and time pressure all shift your style — you may well be direct with a peer and accommodating with a manager in the same week. Anyone who describes their style with one fixed adjective has probably only sampled one context.
What is conflict repair and why does it matter more than style?
Repair is returning to a finished argument, acknowledging it happened and putting it to rest. It matters more than what you did during the argument because stored arguments get reactivated by the next unrelated incident — so a relationship can break over something trivial long after a serious fight that was never closed.
References
- De Dreu, C. K. W., & Weingart, L. R. (2003). Task versus relationship conflict, team performance, and team member satisfaction: A meta-analysis. Journal of Applied Psychology, 88(4), 741–749.
- Jehn, K. A. (1995). A multimethod examination of the benefits and detriments of intragroup conflict. Administrative Science Quarterly, 40(2), 256–282.
- Rahim, M. A. (2002). Toward a theory of managing organizational conflict. International Journal of Conflict Management, 13(3), 206–235.
- Gottman, J. M., & Levenson, R. W. (1992). Marital processes predictive of later dissolution: Behavior, physiology, and health. Journal of Personality and Social Psychology, 63(2), 221–233.
- Gable, S. L., Reis, H. T., Impett, E. A., & Asher, E. R. (2004). What do you do when things go right? The intrapersonal and interpersonal benefits of sharing positive events. Journal of Personality and Social Psychology, 87(2), 228–245.