Interdisciplinary research is a mode of research that integrates two or more bodies of specialized knowledge, to solve problems whose solutions are beyond the scope of a single discipline. This form of research has become an important issue in doctoral education to prepare new generations of scientists to address complex real-world problems. European doctoral education has been subject to policy reforms resulting in new forms of doctoral education. This thesis provides a comprehensive analysis of the doctoral interdisciplinarity in three different kinds of programs in computer science: a traditional doctoral program and two structured programs. This is achieved with three interconnected approaches that aim to measure, understand and assess doctoral interdisciplinarity. Firstly, a bibliometric method is used to measure the interdisciplinarity of publications. The Rao-Stirling diversity index quantifies the interdisciplinarity of a publication based on the disciplines that are integrated into it through its references. Therefore, this approach necessitates the categorization of all references into disciplinary fields, which is a prerequisite rarely fulfilled. As a methodological contribution, this thesis proposes an extension of the index, based on discrete and continuous optimization as well as graph-based pruning, in order to acknowledge the inaccuracies introduced in the measurement by missing bibliographic data. This bibliometric method is subsequently utilized for measuring the interdisciplinarity of doctoral researchers in the three doctoral programs where this investigation is conducted. The second approach aims to understand how and why doctoral researchers conduct interdisciplinary research. Doctoral researchers identified as interdisciplinary through the bibliometric analysis participated in semi-structured interviews. The analysis of their accounts reveals that doctoral interdisciplinarity depends on far more than on explicit strategies implemented to facilitate interdisciplinarity. Personal attributes that emerge prior to the start of doctoral studies influence doctoral researchers' inclination to conduct research in one or multiple disciplines. Additionally, policy and structural factors as well as collaboration processes also influence the degree of disciplines integration conducted by doctoral researchers. These findings lead to the identification of patterns of doctoral interdisciplinarity which contribute to the theory of interdisciplinary education. The third approach assesses how the fulfillment and importance of doctoral interdisciplinarity is perceived by different academic stakeholders. This approach utilizes the 360-degree feedback survey methodology to gather their assessment of factors and processes relevant for interdisciplinarity, which are selected based on a literature review and interviews with interdisciplinary doctoral researchers and professors. Their answers are analyzed using statistical methods in order to investigate the alignment of opinions between groups of stakeholders as well as single-disciplinary and interdisciplinary doctoral researchers. The findings provide useful insights for identifying discrepancies between stakeholders, revealing problematic issues and suggesting actions for improvement. The three approaches (and their respective substantial amounts of information) are mutually complementary and validatory, and therefore strongly corroborating the findings of the investigation. This thesis also contributes to raising the awareness on facilitators and obstacles to doctoral interdisciplinarity and offers recommendations for supporting young interdisciplinary scientists.