Conflicts at Work Derail German Apprenticeships, Panel Study Finds

Apprentices in Germany who report frequent conflicts with colleagues and supervisors are substantially less likely to complete their vocational training, and those who experience both high conflict and little support face the worst outcomes, according to a new study published in Empirical Research in Vocational Education and Training. Using nationally representative panel data, Markus Weißmann and Tobias Roth show that the social climate inside training companies shapes not only whether apprentices finish their programs but also whether graduates end up working for the firm that trained them.

Dual vocational training, which combines on-the-job instruction in a company with school-based teaching, is a cornerstone of the German labor market. More than 40 percent of young adults aged 20 to 25 held a dual vocational training qualification in 2021. Yet the system leaks: around 30 percent of training contracts were terminated prematurely in 2022, and about 24 percent of successful graduates do not start regular employment in their training company. Understanding what drives these dropouts and departures matters both for young people’s careers and for companies, for whom in-house training is the most common way of recruiting skilled workers.

Previous research on training termination relied heavily on interviews with those who had already dropped out. That approach has a blind spot: it excludes the roughly three out of four trainees who complete their training, some of whom surely experienced conflicts too. Retrospective interviews also invite what the authors call retrospective justification, in which former trainees may misattribute responsibility for a failed apprenticeship to companies, supervisors, or colleagues. The new study avoids these pitfalls by measuring the workplace situation while training was still ongoing, before success or failure was known.

The analysis draws on Starting Cohort 4 of the German National Educational Panel Study (NEPS SC4), a nationally representative panel that began with ninth-graders randomly sampled from schools across Germany in the fall of 2010 and interviewed them once or twice a year thereafter. From 16,379 respondents, the researchers identified 7,373 apprenticeship episodes, narrowing to 6,829 episodes from 5,889 respondents after excluding special-needs schools. The final estimation samples used 5,755 first apprenticeship episodes. Support was measured by agreement with the statement that colleagues and superiors take the time to explain new tasks, and conflict by agreement with the statement that the respondent often has problems or conflicts with colleagues and superiors, each grouped into three levels.

The descriptive picture alone is striking. Only ten percent of apprentices reported high levels of conflict, but around 20 percent reported low levels of support, and only 43 percent of trainees reported high support overall. In the sample, around 21 percent of apprenticeships ended prematurely, roughly 80 percent of graduates received an employment offer from their company, about 83 percent of those offers were accepted, and around 66 percent of all graduates started working in their training firm.

The multivariate results, based on logistic regressions reporting average marginal effects with extensive controls for ability, personality traits, socioeconomic background, and training characteristics, show that conflict and lacking support each substantially reduce the likelihood of completing training. Apprentices reporting high conflict had an eight percentage point higher probability of premature termination than those reporting low conflict, and those with medium conflict seven points higher. Low support raised the termination probability by ten percentage points compared with high support. In a combined analysis, apprentices with high conflict and low support had a 15 percentage point lower probability of training success than those with high support and low conflict.

One theoretical expectation failed. The authors had hypothesized that high support might buffer the negative effects of conflict, but the data showed no such buffering. In fact, the association between increasing conflict and lower training success was weakest when support was low and strongest when support was high, the opposite of the buffering pattern.

The study then split the question of staying with the training company into two decisions: whether the company offers the graduate a job, and whether the graduate accepts. High conflict reduced the probability of receiving an employment offer by around six percentage points, a sizable effect that was statistically significant at the ten percent level in the stricter model. Support levels played only a minor role in companies’ offer decisions. By contrast, neither conflict nor support was associated with whether graduates who received offers accepted them, contradicting the researchers’ hypotheses. Only when the two stages were combined did a clear pattern emerge: graduates with high conflict and low support during training were 13 percentage points less likely to remain in their training company, a statistically significant difference. The detailed analyses suggest this is mainly because companies withhold offers rather than because graduates refuse them.

The authors are careful about interpretation. The observational design does not permit causal claims, and panel attrition cannot be ruled out, although the distribution of outcomes closely matches official statistics, suggesting selective attrition is unlikely to bias the results substantially. For some apprentices, the workplace situation was measured only after they had already dropped out, requiring imputation. The data also lack detail on the reasons for and types of conflicts, or on support from different actors such as supervisors versus other trainees.

Despite these limitations, the findings carry practical weight in a country facing skilled labor shortages. The authors suggest that improving interpersonal relationships during training, through conflict prevention measures such as recurring feedback meetings, mentoring programs, or training for instructors, could help secure the long-term engagement of apprentices, and that vocational schools and policy programs could play a stronger role in workplace conflict management.

Subject of Research: How workplace conflict and support during German dual apprenticeships relate to training completion and subsequent employment in the training company

Article Title: Conflict and support during apprenticeship training in Germany: consequences for apprenticeship success and further employment in the company

Article References: Weißmann, M., & Roth, T. (2025). Conflict and support during apprenticeship training in Germany: consequences for apprenticeship success and further employment in the company. Empirical Research in Vocational Education and Training, 17(1), Article 27. https://doi.org/10.1186/s40461-025-00201-8

Image Credits: AI Generated

DOI: 10.1186/s40461-025-00201-8

Keywords: apprenticeship, vocational training, Germany, workplace conflict, social support, NEPS, panel study, training dropout, job offers, labor market transition, dual system, skilled labor