Background <p>Given the importance of culturally sensitive psychotherapy, it is necessary to determine how to promote training in transcultural competence. Transcultural competence consists of several dimensions (skills, knowledge, awareness, therapeutic relationship) that are interrelated, but it is currently unknown how their interrelations facilitate transcultural competence.</p> Methods <p>We conducted a Bayesian network intervention analysis with data from a randomized controlled trial of a web-based training in transcultural competence for clinical psychologists (<i>N</i> = 335). Three networks were calculated to investigate (1) the effect of training on dimensions of transcultural competence and (2) how these dimensions interact with each other. The network included subscales of the Multicultural Counseling Inventory and training allocation as a node, estimated at pre-training, post-training, and two-month follow-up.</p> Results <p>We found strong evidence that training was directly associated with stronger improvement in participants' knowledge, both at post-training and follow-up. In terms of the interplay between the dimensions, weak and strong evidence for associations between skills, knowledge, awareness, and therapeutic relationship was shown across all three time points. We found weak and strong evidence that skills were related to stronger improvement in therapeutic relationship at post-training and at follow-up, respectively.</p> Conclusions <p>While training showed direct effects on knowledge, we found no evidence that it directly improved skills, awareness, or the therapeutic relationship. Still, skills of practicing in a transculturally competent manner seem to play an important role for building therapeutic relationship. The results of this study suggest that the network approach offers new theoretically and empirically grounded opportunities for the evaluation of training in transcultural competence.</p>