Teamwork quality, conflict and psychological safety: what the team data actually supports

A module that teaches a group project has to say something about teamwork, and the evidence is uneven. This is the eighth of ten units, and it asks what the teamwork literature can support about the group the mark describes. The headline claim holds: a six-facet measure of team collaboration was associated with performance across 145 German software teams (Hoegl and Gemuenden, 2001). What is built on top of it is weaker. The same team is rated differently by members, leaders and managers (Hoegl and Gemuenden, 2001; Lindsjorn et al., 2016). The mechanisms it names come mainly from workplace teams whose conditions a student group does not meet (Hackman, 2002). Read the team data as a directional instrument with a marked boundary: trust the direction, distrust the precision, name the observer.

Teamwork quality is a measured construct, and it predicts success

Hoegl and Gemuenden (2001) asked whether teamwork can be measured and answered with a six-facet construct, teamwork quality. Its facets are communication, coordination, balance of member contributions, mutual support, effort and cohesion. Member-rated teamwork quality came from 575 members, leaders and managers of 145 German software teams. It was significantly associated with performance as rated by members, team leaders and team-external managers, and strongly with work satisfaction. Its facets name different work: coordination and balance of contribution sit inside the same measure as cohesion. A group attending only to whether it feels agreeable reads one facet of six.

The same team is rated differently by different observers

That association varied in strength with the perspective of the performance rater (Hoegl and Gemuenden, 2001). Lindsjorn and colleagues (2016) reproduced the pattern in software. Across 477 respondents in 71 agile teams in 26 companies, teamwork quality was positively related to performance when members and team leaders rated it, and negligibly when product owners rated it. Members also rated the effect on their learning and satisfaction as strongly positive. Any team-health score, and any contribution mark assembled from team ratings, carries whoever produced it: the observers are not noise to average away but measure partly different things.

The textbook claim about task conflict does not survive, and the process is where the leverage sits

The conflict literature is where confident teaching is most easily embarrassed. Jehn (1995) examined 105 work groups and management teams and drew the distinction the field now uses: task conflict, disagreement about the work, against relationship conflict, friction between people. In routine groups, task disagreement was detrimental, and in non-routine groups it was harmless or helpful, which is where the textbook claim comes from; both types damaged members' satisfaction, liking of teammates and intent to remain. Norms encouraging open discussion brought more relationship conflict without improving how constructively anyone handled it. De Dreu and Weingart (2003) then meta-analysed the field: relationship conflict was strongly and negatively related to performance and satisfaction, and task conflict was also negatively related, against the textbooks, with conflict more damaging in complex than in production tasks. A team that adopts an open-disagreement rule is therefore not buying the mechanism the textbook names.

If conflict type does not carry the effect, the process might. Behfar and colleagues (2008) studied 57 autonomous teams. Groups that maintained top performance shared three conflict-resolution tendencies: focusing on content rather than delivery style, stating the reasons behind decisions about distributing work, and assigning work by task expertise rather than by volunteering or convenience. Tjosvold (2007) reaches a compatible position. Conflict managed cooperatively can make teamwork effective, competitive orientations produce avoidance and escalation, and the field confounded conflict with competition by treating type, not management, as decisive.

Psychological safety is a condition for learning, not a synonym for comfort

Edmondson (1999) introduced team psychological safety as a shared belief that the team is safe for interpersonal risk taking and tested it on 51 work teams in a manufacturing company. It was associated with learning behaviour, which mediated the link to performance, while team efficacy was not. The design carries that claim, because antecedent, process and outcome variables were measured across the same 51 teams. A survey that asked only about comfort would report the first half of the chain and stop. Edmondson and Lei (2014) review the construct from its 1960s origins through its renaissance and distinguish it from its neighbours, trust and cohesion among them.

The constructs travel in direction, and a student project meets few of the conditions

None of these teams is the group in the room. Hackman (2002) argues that effectiveness rests on five conditions: a real team, a compelling direction, an enabling structure, a supportive organisational context and expert coaching. A student project group is time-limited, assigned rather than self-selected, directed by a brief and a marking scheme, and outside any team reward system, so several conditions are thin. Its members also carry different availability: some are carers, some commute. The constructs do travel in direction, though. Lindsjorn and colleagues (2016) found teamwork quality no higher in agile teams than in comparable traditional ones. Alami, Zahedi and Krancher (2024) combined 20 interviews with a survey of 423 respondents, finding psychological safety supported admitting mistakes and initiatives bearing on software quality. Verwijs and Russo (2024) surveyed 1,118 participants in 161 software teams, finding that age diversity helped effectiveness while gender diversity related to relational conflict, and that psychological safety contributed directly to teamwork without moderating the diversity-to-effectiveness link. Direction replicates; precision does not.

One study has now been run on student teams themselves. Ratiu and colleagues (2025) collected time-lagged data from 1,002 students in 199 project teams: contribution conflict lowered team performance indirectly through psychological safety, and exploring different perspectives did not improve performance on its own. The design is self-report and single-country, so the gap narrows rather than closes.

The direction and the levers survive the boundary, and nothing more

Two limits should travel with any team measure into a project. The first is the observer: the same construct scores differently by role, and in software the product owner saw no effect where members and leaders saw one (Lindsjorn et al., 2016). The second is measurement itself, which the field's fifty-year review still lists among its open challenges (Salas et al., 2008). What survives is the direction and three levers.

Coordination translates most cleanly. Marks, Mathieu and Zaccaro (2001) define team process within a multiphase episodic framework and sort the dimensions into transition, action and interpersonal processes. The consequence is timing: planning belongs between episodes of work. Coordination is a schedule, not a personality.

Disagreement handled through a process addressing content, reasons and expertise is the second lever (Behfar et al., 2008). Attention to safety is the third, because safety predicts learning (Edmondson, 1999; Alami et al., 2024).

Contested question: is task conflict good or bad for a team?

The contested question is whether disagreement about the work helps or harms. The conflict-positive position holds that handling decides the outcome. Jehn (1995) found task disagreement harmless and sometimes helpful in non-routine work, and Tjosvold (2007) argues that cooperative handling turns conflict into a resource while competition produces avoidance and escalation. The meta-analytic position holds that the sign is negative whatever the theory says:

In contrast to what has been suggested in both academic research and introductory textbooks, however, results also revealed strong and negative ... correlations between task conflict, team performance, and team member satisfaction. (De Dreu and Weingart, 2003)

Behfar and colleagues (2008) occupy the middle ground, locating the difference in resolution tendencies rather than conflict type. What would settle it is a study that varies only the conflict-management strategy while holding task type constant. It would measure conflict type and management together, and report their correlation, since task conflict was less harmful when weakly correlated with relationship conflict (De Dreu and Weingart, 2003). No such study exists at student-team scale, and the field evidence is correlational.

Rate the two cases by kind. The conflict-positive case is one field study of 105 groups and a theory (Jehn, 1995; Tjosvold, 2007). The meta-analytic case is a synthesis of the whole field (De Dreu and Weingart, 2003). The middle position is a study of 57 teams (Behfar et al., 2008). The strongest design sits with the meta-analysis, which reports a negative sign. While the dispute is open, treat conflict type as a poor predictor and manage the process: address content rather than delivery style, state the reasons for a decision about work, and assign by expertise. The strongest objection is that every lever here was measured once, in teams whose conditions a student group does not meet (Hackman, 2002), and a survey of 1,118 participants shows the direction itself is conditional (Verwijs and Russo, 2024).

Where the argument is heading

The direction of travel is software-specific and increasingly sceptical of single measures. Alami and colleagues (2024) connect psychological safety to quality-related behaviour in agile teams. Verwijs and Russo (2024) show the diversity and conflict claims are more conditional than the general models imply, so friction is a reason to design the process, not to compose teams defensively. For a student team the reading is stable and modest. The mechanisms are best used as things a team does at named points, not as a score. What the literature still does not supply is a comparison of those student teams against workplace teams on the same measures.

The sources this unit rests on

• Hoegl, M. and Gemuenden, H. G. (2001). Teamwork quality and the success of innovative projects: a theoretical concept and empirical evidence. Organization Science, 12(4), 435-449. doi:10.1287/orsc.12.4.435.10635 Tested six facets of teamwork quality on 575 members, leaders and managers of 145 German software teams. Member-rated quality predicted performance as rated by all three observer groups, though the magnitude varied by rater.
• Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350-383. doi:10.2307/2666999 Tested a model of team learning on 51 work teams in a manufacturing company. Psychological safety was associated with learning behaviour, which mediated the link to performance, while team efficacy was not.
• Edmondson, A. C. and Lei, Z. (2014). Psychological safety: the history, renaissance, and future of an interpersonal construct. Annual Review of Organizational Psychology and Organizational Behavior, 1(1), 23-43. doi:10.1146/annurev-orgpsych-031413-091305 Reviews the construct from its 1960s origins through its renaissance and maps what it predicts. It distinguishes safety from its neighbours, trust and cohesion among them.
• Jehn, K. A. (1995). A multimethod examination of the benefits and detriments of intragroup conflict. Administrative Science Quarterly, 40(2), 256-282. doi:10.2307/2393638 Examined 105 work groups and management teams and separated task from relationship conflict. Task disagreement was detrimental in routine work and sometimes beneficial in non-routine work, and both types damaged satisfaction, liking and intent to remain.
• De Dreu, C. K. W. and 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. doi:10.1037/0021-9010.88.4.741 Meta-analysed the conflict literature. Relationship conflict was strongly and negatively related to performance and satisfaction, and task conflict also negatively related, against the textbooks, with conflict more damaging in complex tasks.
• Marks, M. A., Mathieu, J. E. and Zaccaro, S. J. (2001). A temporally based framework and taxonomy of team processes. Academy of Management Review, 26(3), 356-376. doi:10.5465/amr.2001.4845785 Defined team process within a multiphase episodic framework and sorted process dimensions into transition, action and interpersonal processes. It supplies the unit's timing rule for coordination.
• Salas, E., Cooke, N. J. and Rosen, M. A. (2008). On teams, teamwork, and team performance: discoveries and developments. Human Factors: The Journal of the Human Factors and Ergonomics Society, 50(3), 540-547. doi:10.1518/001872008X288457 Surveyed fifty years of team-performance research and set out eight discoveries and five challenges, placing naturalistic study and better measurement among the challenges. It marks which team claims remain open problems.
• Hackman, J. R. (2002). Leading teams: setting the stage for great performances. Boston, MA: Harvard Business School Press. https://archive.org/details/leadingteams00jric Argues that effectiveness rests on five conditions: a real team, a compelling direction, an enabling structure, a supportive organisational context and expert coaching. It names the conditions a time-limited student group lacks.
• Behfar, K. J., Peterson, R. S., Mannix, E. A. and Trochim, W. M. K. (2008). The critical role of conflict resolution in teams: a close look at the links between conflict type, conflict management strategies, and team outcomes. Journal of Applied Psychology, 93(1), 170-188. doi:10.1037/0021-9010.93.1.170 Studied 57 autonomous teams and found that top performers shared three conflict-resolution tendencies, including focusing on content over delivery and assigning work by task expertise. It locates the effect in process rather than conflict type.
• Tjosvold, D. (2007). The conflict-positive organization: it depends upon us. Journal of Organizational Behavior, 29(1), 19-28. doi:10.1002/job.473 Argues that conflict managed cooperatively can make teamwork effective while competitive orientations produce avoidance and escalation. It holds that the field confounded conflict with competition and overrated the importance of conflict type.
• Lindsjorn, Y., Sjoberg, D. I. K., Dingsoyr, T., Bergersen, G. R. and Dyba, T. (2016). Teamwork quality and project success in software development: a survey of agile development teams. Journal of Systems and Software, 122, 274-286. doi:10.1016/j.jss.2016.09.028 Surveyed 477 respondents in 71 agile teams across 26 companies. Teamwork quality predicted performance when members and leaders rated it, negligibly when product owners did, and was no higher than in traditional teams.
• Alami, A., Zahedi, M. and Krancher, O. (2024). The role of psychological safety in promoting software quality in agile teams. Empirical Software Engineering, 29(5). doi:10.1007/s10664-024-10512-1 Combined 20 interviews with a survey of 423 respondents. Psychological safety in agile teams supported admitting mistakes and taking initiatives that bear on software quality.
• Verwijs, C. and Russo, D. (2024). The double-edged sword of diversity: how diversity, conflict, and psychological safety impact software teams. IEEE Transactions on Software Engineering, 50(1), 141-157. doi:10.1109/tse.2023.3339881 Surveyed 1,118 participants in 161 software teams. Age diversity helped team effectiveness while gender diversity related to relational conflict, and psychological safety contributed directly to teamwork and less conflict but did not moderate the diversity-to-effectiveness link.
• Ratiu, L., Buzea, I. M., Rus, C. L. and Otoiu, C. (2025). Beyond the surface: unravelling the relationships between conflict, psychological safety, team learning, and performance in student teams. Studies in Higher Education, 51(6), 1338-1353. doi:10.1080/03075079.2025.2509148 Time-lagged study of 1,002 students in 199 project teams; contribution conflict lowered psychological safety, which mediated team performance. The rare student-team case; its limit is self-report data from one country.