Abstract
Kazakhstan's doctoral education policy requires all PhD students to have both a local and a foreign supervisor. This co-supervision model was designed to connect doctoral training with international academic standards. Little empirical work has asked what this arrangement means for students' experience of supervision, or whether satisfaction with feedback from each supervisor type tracks internationalization outcomes. Survey data were collected from 845 doctoral students and graduates across institutions in Kazakhstan. Valid sample sizes varied across analyses, ranging from 586 to 828 cases, because the satisfaction items included response options such as ‘not applicable’ and ‘prefer not to answer’. Independent samples t-tests, chi-square tests, and Spearman rank-order correlations were used to test associations between supervisor feedback satisfaction and three outcomes: Scopus publication output, international conference participation, and international internship outcomes. Satisfaction with foreign supervisor feedback predicted higher Scopus output (t = 3.06, p = .002) and conference participation (χ² = 3.85, p = .049), and correlated moderately with internship outcomes (ρ = .56). For local supervisor feedback, no comparable association appeared with publication output or conference attendance; only the internship correlation reached significance, and at a notably weaker magnitude (ρ = .37). Internationalization in Kazakhstan's PhD system, these results indicate, runs mainly through the foreign half of the co-supervision arrangement. The local half contributes far less to the outcomes the policy was designed to produce from both sides. What the model needs is not a new mandate but the institutional conditions, reduced teaching loads and dedicated mobility funding chief among them, that would let local supervisors operate on comparable terms.