摘要
Introduction: Policy entrepreneurship, widely studied in public policy and administration, has traditionally focused on high-ranking decision-makers or street-level bureaucrats in developed nations. This research, however, shifts attention to mid-level managers in developing countries, where institutional fragility presents distinct challenges and opportunities. It posits that, in such contexts, policy entrepreneurship often serves as a complementary mechanism to institutional frameworks, potentially enhancing the impact of policy entrepreneurs. By employing Q-methodology, this study investigates how mid-level government managers in developing countries perceive policy entrepreneurship, addressing a notable gap in the literature, which has predominantly centred on developed countries or different governance levels. Through an integration of natural language processing with Q-methodology, it examines the subjective perspectives of thirty-five mid-level government managers. Methodology: This study employs Q-methodology, a tool for examining subjectivity, in conjunction with natural language processing techniques to generate Q-statements. Traditionally, Q-methodology has relied on the researcher’s judgment to identify the Q-set, drawing from sources such as journal articles, newspapers, and interviews. However, this conventional approach risks limiting the comprehensiveness of the concourse universe. While the size of the P-set (n) is less critical in Q-methodology, ensuring the thoroughness of the Q-set is essential for achieving its objectives. By integrating natural language processing, this approach offers a more comprehensive understanding of mid-level government managers’ attitudes and perceptions of policy entrepreneurship. Findings: The findings reveal four distinct perspectives among respondents: ‘External Innovators,’ who believe in external sources of policy innovation; ‘Strategic Orchestrators,’ emphasising the importance of strategic, individual-driven policy design; ‘Political Pragmatists,’ advocating for the efficacy of politicians in promoting policies; and ‘Elite Sceptics,’ who are critical of policy entrepreneurs’ motivation. These factors highlight the complex and divided views of policy entrepreneurship among mid-level government managers in developing countries. Conclusion: This study contributes to this field by expanding our understanding of policy entrepreneurship in new demographic and geographic contexts. It underscores the utility of Q-methodology in public policy research and provides insights that can inform capacity-building efforts and strategies for effective policy implementation in developing countries. These findings also suggest avenues for future research, emphasising the need for more comprehensive studies on the role and perceptions of mid-level managers in the policy process.