Case Study 2: The Making of the TKI — Research, Prevalence, and What Organizations Learned

Overview

This case study examines the historical and scientific foundation of the Thomas-Kilmann Conflict Mode Instrument (TKI). We explore the problem Kenneth Thomas and Ralph Kilmann were trying to solve, the theoretical lineage of their framework, what the instrument actually measures and how it was validated, and what more than four decades of organizational research has revealed about conflict style prevalence — including consistent findings around gender, culture, organizational role, and the relationship between conflict style and leadership effectiveness.

The Problem That Needed Solving: Organizations in the Early 1970s

By the late 1960s and early 1970s, organizational psychology was confronting a practical puzzle. Management consultants and organizational theorists were increasingly certain that conflict was a significant driver of organizational dysfunction — lost productivity, failed teams, leadership derailment, wasted resources in adversarial negotiations — but they lacked a clean, measurable way to describe how different people handled it.

The dominant frameworks of the time were useful but incomplete. Win-lose versus win-win thinking (popularized in the negotiation literature) was a start, but it described outcomes rather than behaviors. Personality typologies gave general tendencies but lacked specificity about conflict behavior in particular. And clinical psychology's frameworks, rooted in therapeutic contexts, didn't translate readily to workplace applications.

The practical problem was vivid: organizational development consultants working with management teams could see that people in the same room, facing the same conflict, responded radically differently — some pushed hard, some withdrew, some found creative solutions, some gave up their own position entirely. These differences were consistent enough that they seemed like something you could measure. But no validated instrument existed to do it.

Kenneth Thomas, then an organizational behavior scholar at UCLA (later at the Naval Postgraduate School), and Ralph Kilmann, then at the University of Pittsburgh (later widely known for his organization transformation work), set out to build one.

The Theoretical Lineage: Blake and Mouton's Conflict Grid

Thomas and Kilmann did not begin from scratch. They drew substantially on the work of Robert Blake and Jane Mouton, two organizational psychologists who had, in the early 1960s, developed what they called the "Conflict Grid" (or "Managerial Grid" in a slightly different formulation). Blake and Mouton had proposed that managerial behavior in conflict situations could be mapped on two axes: concern for production and concern for people. These two dimensions produced five behavioral zones that looked, with adjustment, very similar to what the TKI would ultimately describe.

Thomas and Kilmann refined and operationalized this framework in several important ways. First, they re-labeled the two dimensions in terms that applied more broadly than a managerial context: instead of "concern for production" and "concern for people," they used "assertiveness" (concern for one's own interests) and "cooperativeness" (concern for others' interests). This reframing made the model applicable across settings — workplace, family, interpersonal — rather than only to managerial relationships.

Second, and crucially, they developed a forced-choice assessment instrument to actually measure the five modes: a series of paired statements in which respondents must choose between two options that each reflect a different conflict behavior. This forced-choice format was designed to reduce social desirability bias — the tendency to answer how one thinks one should behave rather than how one actually does behave. By requiring respondents to choose between two equally positive-sounding behaviors rather than rating how much they agreed with statements, the instrument pushes for more accurate self-report.

The original TKI was published in 1974. It has since been revised (most significantly in 2007) to update language, improve psychometric properties, and expand its comparative norms database. The 2007 revision also introduced percentile scores by work role, providing organizations with more nuanced benchmarks for understanding a given person's style relative to others in similar positions.

What the TKI Measures — and How

The TKI presents respondents with thirty pairs of statements. For each pair, respondents must choose which of the two statements better describes their behavior. Neither statement in a pair is obviously "better" — both reflect legitimate behaviors. This is the forced-choice format at work.

For example, a pair might offer something like: - A: I try to find a compromise solution. - B: I attempt to deal with all of his and my concerns.

Or: - A: I sometimes avoid taking positions that would create controversy. - B: I let the other fellow have some of his points if he lets me have some of mine.

Responses are scored across the five mode dimensions, producing a raw score and — in the revised instrument — a percentile ranking relative to a large normative sample. High scores on any given mode do not mean the respondent always uses that mode; they mean the respondent uses it more frequently than the comparative norm group. Low scores indicate underuse relative to the norm.

This scoring approach makes an important theoretical point: conflict modes are not categorical (you are or aren't an avoider) but dimensional (you avoid more or less than other people in comparable situations). The instrument doesn't produce a type; it produces a profile — a distribution across five modes that can be compared to norms, tracked over time, and discussed in coaching and team contexts.

What the TKI Does Not Measure

It is worth being explicit about the TKI's limits. The instrument:

Does not predict behavior in specific high-stakes situations. It measures habitual tendencies under ordinary conditions. Under acute stress, many people shift dramatically toward competing or avoiding regardless of their assessed profile.
Does not assess emotional intelligence or relational skill. Two people can both score high on collaborating, but one may be a skilled and empathetic collaborator while the other approaches collaboration as an intellectual exercise without genuine attunement to the other party.
Is a self-report instrument, subject to self-knowledge limits. Even with forced-choice design, respondents may not have accurate insight into how they actually behave. As the chapter notes, competing people often perceive themselves as "just being direct"; accommodating people often perceive themselves as "keeping the peace."
Does not account for domain variation. The TKI typically asks respondents to answer based on general work situations. As we have seen, people often use different modes in different relational domains.

These limits don't diminish the instrument's value; they contextualize it. The TKI is most powerful not as a definitive personality classification but as a tool for reflection, conversation, and intentional development.

What the Research Reveals: Prevalence Across Populations

Four decades of TKI data have produced a substantial research base on how conflict modes are distributed across populations. The findings are nuanced and sometimes surprising.

Which Modes Are Most Common?

Large-scale analyses of TKI norms consistently find that avoiding and accommodating are the most overused modes in organizational settings, while collaborating and competing are the most underused — particularly in management and professional populations.

This finding counters the intuitive assumption that competitive, hard-driving behavior is the default in organizational life. In reality, the data suggest that most people in most organizational situations err on the side of conflict suppression rather than engagement. Avoidance is particularly prevalent: across the normative database, avoiding is the most common response to conflict situations across nearly all demographic groups.

The dominance of avoidance in organizational settings has important implications. Organizations where most people avoid most conflict are not necessarily peaceful places — they are often places where chronic, unaddressed problems accumulate below the surface, surfacing only in crisis or in the subtle form of disengagement, turnover, and burnout.

Collaborating is among the least common modes in practice, despite being the mode most people describe as their ideal when asked abstractly how they want to handle conflict. This gap between aspirational and actual behavior is one of the most consistent findings in conflict research, and it has important implications for how organizations design meetings, decision-making processes, and feedback systems. Saying you prefer collaboration is not the same as having the skills and conditions to actually do it.

Gender Differences

The TKI research consistently shows gender differences in conflict mode distribution, with men scoring higher on competing and women scoring higher on accommodating, on average. These findings replicate across studies conducted across multiple decades and countries.

Thomas and Kilmann, and the researchers who have extended their work, are careful to note what these findings mean and what they don't:

What they mean: There are real, consistent average differences in how men and women in organizational settings report approaching conflict. These differences are large enough to be statistically meaningful.

What they don't mean: These are not biological differences. They reflect differential socialization — specifically, the different messages men and women receive about what conflict behavior is acceptable and what it signals about character and role. Men in many organizational cultures are implicitly rewarded for competitive directness (it reads as "decisive leadership") and penalized for appearing too accommodating (which may read as "weak"). Women in the same cultures often experience the reverse: accommodation and relational attunement are rewarded while competing is penalized (the notorious "she's too aggressive" feedback that has no male equivalent).

These socialization pressures are neither universal nor permanent. Research on gender and conflict style in organizations has found that the gap narrows significantly when women hold high-authority positions — and some studies find that women in senior leadership show competing scores comparable to their male counterparts. This suggests that the difference in default style is substantially explained by positional factors and organizational culture, not by fixed traits.

Deborah Tannen's research on gender and communication (You Just Don't Understand, 1990; Talking from 9 to 5, 1994) provides important qualitative texture to these quantitative findings. Tannen's argument — that men and women are often socialized into different conversational "frames" around status and connection — helps explain not just what the TKI data shows but why those differences arise and how they play out in real conversations.

Cultural Differences

The TKI has been administered across more than forty countries since its development, and the cross-cultural findings are among the most theoretically rich in the literature. They complicate the assumption that the TKI's two dimensions (assertiveness and cooperativeness) are culturally universal.

Collectivist versus individualist cultures: Research comparing conflict styles in collectivist cultures (where group harmony, face-saving, and interdependence are primary values) and individualist cultures (where personal goals, directness, and independence are more valued) consistently finds that collectivist-oriented respondents score higher on avoiding and accommodating. This is not a finding of "avoidance as pathology" in these populations; it reflects a coherent value system in which preserving social harmony and protecting face — for oneself and others — is understood as a form of relational intelligence, not conflict failure.

This finding matters enormously for organizations operating across cultural contexts. A team member who consistently "doesn't raise issues" may not be an avoider in the clinical sense; they may be operating from a cultural framework in which directly voicing disagreement with a superior is considered disrespectful rather than courageous. Understanding this distinction changes how managers should interpret conflict style data and design team communication norms.

East Asian research contexts: Studies of conflict style in Japan, China, South Korea, and Taiwan have generally found lower competing scores and higher avoiding/accommodating scores compared to North American and Northern European norms. Researchers attribute this to both Confucian values around hierarchy and harmony and to the face-saving dynamics that structure many East Asian professional interactions.

Latin American and Mediterranean contexts: Research here is more mixed, often finding higher scores on what might be called relational assertiveness — direct emotional expression within relationships — combined with strong face-preservation norms in formal or hierarchical contexts. This produces conflict patterns that don't map cleanly onto North American TKI norms.

Implication for practitioners: Using North American TKI norms to interpret the scores of people from other cultural backgrounds is methodologically problematic. The revised TKI (2007) expanded its normative database to include more diverse populations, but cultural interpretation of conflict style data remains an area requiring significant professional judgment.

Organizational Role and Level

Consistent findings across TKI research show that conflict mode distribution correlates with organizational role and level in predictable ways:

Senior leaders tend to score higher on competing than middle managers or individual contributors. This could reflect self-selection (people with competing tendencies are more likely to seek or be promoted to leadership roles) or role modeling (competing is implicitly validated at the top).
HR professionals and people in counseling, mediation, and coaching roles tend to score higher on collaborating and lower on competing than the general management population.
People in engineering, technical, and quantitative roles in some studies show higher avoiding scores — a finding researchers attribute partly to the cultural norms of those fields, where conflict can be seen as non-technical and therefore outside professional scope.
Sales professionals often show higher competing scores, reflecting both self-selection and the role's inherent win-lose orientation.

These role-level findings have practical implications for team composition. A team of all avoiders will likely suppress important disagreements. A team of all competitors may win arguments but struggle to make lasting, committed decisions. Understanding the conflict style distribution of a team helps leaders anticipate how that team will handle pressure, ambiguity, and disagreement — and design processes accordingly.

What Organizations Learned When They Started Measuring

The organizational story of the TKI is, in part, the story of what happened when companies began systematically assessing and discussing conflict style. Several themes emerge from the consulting and organizational development literature of the 1980s and 1990s, as the TKI moved from academic instrument to widely used management tool.

The Avoidance Discovery

Perhaps the most consistent organizational "aha" moment produced by TKI adoption was the discovery of how thoroughly avoiding dominated organizational conflict behavior — including, and especially, among people who believed themselves to be decisive and direct. Managers who thought of themselves as straightforward communicators were surprised to find that their TKI profiles showed significant avoidance around personnel issues, upward feedback, and interpersonal conflict with peers. The instrument forced a reckoning with the gap between self-perception and behavior.

This discovery had implications for how organizations thought about manager development. Confronting underperformance, giving hard feedback, surfacing disagreement in meetings — these were not "soft skills" that existed outside the core of management; they were conflict behaviors that could be measured and developed.

The Collaboration Gap

Organizations also discovered the collaboration gap: the consistent finding that collaboration was the most widely endorsed value in team culture statements but the least commonly practiced behavior under pressure. When workloads increased, deadlines tightened, and stakes rose — precisely when collaboration would be most valuable — teams reverted to competing and avoiding.

This finding has shaped much of the organizational communication research of the past three decades. It prompted scholars and practitioners to ask not just "how do we teach collaboration?" but "what conditions make collaboration possible or impossible?" The answers pointed toward psychological safety (the belief that one can speak up without punishment), trust developed through consistent experience, and structural supports like adequate time and clear process — all of which can be designed into organizations, not just wished for.

The Team Composition Question

When organizations began administering TKI assessments to teams rather than individuals, they discovered that team-level conflict style distribution predicted team behavior in measurable ways. Teams with high average avoiding scores tended to underperform on tasks requiring surfacing of disagreements and diverse perspectives. Teams with high average competing scores sometimes outperformed on convergent tasks (where decisiveness mattered) but struggled on complex, ambiguous problems requiring genuine integration of multiple views.

The most effective teams, in multiple studies, were those with distributed style profiles — some members tending toward competing, some toward collaborating, a few who could strategically deploy accommodating or compromising — combined with awareness of their own distribution and the flexibility to adjust their collective mode to the task at hand. Style diversity, combined with meta-awareness of that diversity, appeared to be a significant predictor of team effectiveness.

The Leadership Insight

The TKI research produced a counterintuitive finding about leadership that has influenced leadership development practice significantly: the conflict styles most associated with leadership are not the ones most associated with leadership effectiveness.

Competing is the mode most commonly associated with leadership in popular culture and in the self-assessments of many people who rise into leadership roles. But TKI research and adjacent leadership studies consistently find that the most effective leaders — across multiple performance indicators — show higher scores on collaborating and lower scores on avoiding than the average manager. They are not necessarily low on competing; in fact, the ability to compete when appropriate is important. But they don't lead with it.

The implication is that organizations that promote primarily on the basis of competitive dominance — and many do — may be systematically selecting for a conflict style that is not optimal for long-term leadership effectiveness. This finding has been influential in executive coaching, where helping leaders diversify their conflict style repertoire (particularly developing collaborating capacity) has become a core part of the work.

The Instrument Today: Scope and Ongoing Research

The TKI remains one of the most widely used conflict assessments in the world, administered by CPP, Inc. (now part of The Myers-Briggs Company). As of the mid-2010s, an estimated 4 million people had completed the instrument. It is used in business, government, healthcare, education, military, and therapeutic contexts.

Ongoing research has extended the original framework in several directions:

Neuroscience and conflict: Research on the neuroscience of threat response has offered biological grounding for some of the TKI findings — particularly why people revert to fight (competing) or freeze/flee (avoiding/accommodating) under high arousal, and why collaboration is neurologically expensive and requires calm enough conditions to sustain.
Online and remote work: Studies conducted in the 2010s and 2020s have examined how conflict modes shift in virtual team environments. Preliminary findings suggest that avoiding increases in asynchronous communication contexts (email, chat) because the friction of real-time conflict is more easily postponed — and that organizations relying heavily on remote communication need active design interventions to prevent the dominance of avoidance.
Integration with other frameworks: Researchers have examined the intersection of TKI modes with emotional intelligence measures, Big Five personality traits, and leadership competency frameworks, finding meaningful correlations that inform how conflict style fits into broader models of professional development.

Key Takeaways from This Research Perspective

The TKI emerged from a real organizational problem: the lack of a clean, measurable way to describe consistent individual differences in conflict behavior. Thomas and Kilmann built on existing theory (Blake and Mouton) and added empirical rigor through forced-choice instrument design.
The most important finding across four decades of data is also the most counterintuitive: avoidance, not competing, is the most prevalent conflict mode in organizational settings — and it is the one most associated with organizational dysfunction.
Gender and cultural differences in conflict style are real, consistent, and substantially shaped by socialization and context rather than biology. Interpreting these differences requires care: a finding of higher accommodating in one cultural or gender group is not a finding of weakness but of a different value orientation and social permission structure.
The gap between aspirational and actual conflict behavior is one of the most consistent findings in the field. Most people endorse collaboration. Most people, under pressure, avoid or compete.
Team-level conflict style distribution matters as much as individual style, and the most effective teams tend to be both diverse in style and aware of that diversity.
The TKI has influenced leadership development significantly, particularly by challenging the assumption that competitive dominance is the leadership default and demonstrating that collaborative capacity is more consistently associated with leadership effectiveness.

Discussion Questions

The TKI uses a forced-choice format to reduce social desirability bias. Can you think of situations in which even the forced-choice format would fail to prevent people from answering in socially desirable ways? What would make the instrument more or less vulnerable to this problem?
The finding that collaboration is the most endorsed but least practiced mode in organizations under pressure is striking. What organizational conditions do you think would be necessary to close that gap? Have you experienced any organization that seemed to actually practice what it preached about collaborative culture?
The research suggests that promoting leaders primarily on the basis of competitive behavior may systematically select for a conflict style that isn't optimal for leadership effectiveness. If you were designing an organizational promotion process, how would you account for conflict style data in leadership selection?
The cultural findings suggest that conflict style tools developed in North American contexts may not translate cleanly across cultural settings. What are the ethical responsibilities of practitioners who use the TKI in cross-cultural organizational contexts?
The neuroscience research suggests that collaboration is neurologically expensive — it requires calm enough conditions to sustain. What does this imply about when organizations should schedule high-conflict conversations or significant team decisions?