Yuki Shiraito
Associate Professor
Research Interests
Make your next move.
Prepare your application
Review drafts before you send them.
Explore related searches
About
Yuki Shiraito is an Assistant Professor at the University of Michigan, USA. His research areas include public policy perception, immigration policy stances, and the legitimacy of autocratic regimes. He has authored several articles on topics such as international law compliance and machine learning applications in text classification.
Articles (12)
Recent research on international organizations (IOs) has highlighted their capacity to influence state behavior by shaping domestic public opinion, a key channel for policy change. This paper investigates an important extension of this mechanism: whether domestic leaders can proactively leverage IO endorsements to bolster support for contentious policies. Drawing on a survey experiment examining attitudes toward a controversial tax increase in Japan, we test how the effectiveness of this tactic depends on how IOs are perceived across multiple dimensions, including neutrality and expertise. Our results show that referencing IO endorsements reduces resistance to the policy. Crucially, their impact is strongest when the IO is seen as aligned with national interests. In contrast, traits such as impartiality and technical expertise—emphasized in previous studies—play a more limited role. Moreover, the persuasive power of endorsements is concentrated among government supporters, further underscoring the importance of perceived interest alignment as a critical dimension of persuasiveness. These findings offer new insights into when and how IOs shape mass attitudes and call attention to the strategic value of perceived IO biases for domestic elites.
Year:
2025
Studies in political science have revealed that voters evaluate candidates’ policy platforms based on gendered views, where women are expected to handle issues such as education well, while men are perceived to be better at issues such as national security. However, the extent to which voters’ views are gendered on immigration policy is less known, as existing theories offer varying interpretations of whether this issue is more aligned with the feminine or masculine stereotype. This paper empirically examines gendered evaluations of immigration policy platforms by conducting a survey experiment in Japan. Our experimental vignette presents a hypothetical candidate who is affiliated with a traditionally anti-immigration party but supports expanding immigration. We manipulate the gender of the candidate and the gendered framing of the position, and examine their interaction effects on attitudes to the candidate. Our experimental results show that the respondents do not evaluate the candidate based on gender and its interaction with the framing of the policy, suggesting that gender bias in voter evaluations may not be as severe as the literature expects in the immigration policy area.
Year:
2025
Year:
2025
Japan had one of the most restrictive entry bans between 2020 and 2023 to curb the spread of coronavirus disease 2019 (COVID-19). While there was critical coverage of this entry ban by foreign media, how did Japanese citizens perceive these restrictions? We fielded a survey in February 2022 with 6,033 respondents, in which we asked the respondents whether they supported the ban and what types of foreigners they would support getting entry to the country. Our results show that the overall level of support for the ban was very high. Those with a stronger sense of ethnic identity expressed a higher level of support, although those with more of a civic sense of national identity were also supportive of the ban. Furthermore, while the respondents were more likely to tolerate the entry of foreign spouses of Japanese citizens and resident foreigners compared to the other immigrant types, even the supporters for the admission of the spouse category were only barely a majority. In sum, our data show that Japanese citizens were uniformly negative against any entries by non-Japanese citizens into Japan during the COVID-19 pandemic.
Year:
2025
Social scientists often classify text documents to use the resulting labels as an outcome or a predictor in empirical research. Automated text classification has become a standard tool since it requires less human coding. However, scholars still need many human-labeled documents for training. To reduce labeling costs, we propose a new algorithm for text classification that combines a probabilistic model with active learning. The probabilistic model uses both labeled and unlabeled data, and active learning concentrates labeling efforts on difficult documents to classify. Our validation study shows that with few labeled data, the classification performance of our algorithm is comparable to state-of-the-art methods at a fraction of the computational cost. We replicate the results of two published articles with only a small fraction of the original labeled data used in those studies and provide open-source software to implement our method.
Year:
2024
Year:
2023
Year:
2023
Year:
2023
Year:
2023
Year:
2023
Year:
2023
Year:
2023
Collaborators (2)
Guoer Liu
Assistant Professor
University of California, San Diego
Kenneth McElwain
Professor
University of Tokyo
Make your next move.
Prepare your application
Review drafts before you send them.