【論文（査読あり）】
1. “Why Voters Prefer Politicians with Particular Personal Attributes: The Role of Voter Demand for Populists.”
Recent studies in democratic countries suggest that voters generally prefer candidates with specific personal attributes, such as being female and young. However, some of these patterns cannot necessarily be explained by voters’ expectations of candidates’ competence. Building on a growing body of literature that addresses populist attitudes as an important factor influencing voters’ political preferences, this study hypothesizes that candidates’ personal attributes shape voters’ perceptions of their populist attitudes and that such perceptions mediate the relationship between personal attributes and voter preferences. A conjoint experiment conducted in Japan showed that several personal attributes substantially influenced candidates’ perceived anti-elitism and people-centrism. An additional experiment to disentangle causal mechanisms suggested that, albeit inconclusive, young candidates were more likely to be preferred because of voters’ expectations of their populist tendencies. Methodologically, this study illustrates advancements in the conjoint experiment design to elucidate causal mechanisms, with a careful discussion of necessary assumptions.
Political Studies. Accepted.
[replication files]
2. “Party Theory of Parliamentary Debate and the Endogenous Nature of Parliamentary Institutions: Theoretical Implications from Japan’s Diet.”
We add new insights to the party theory of parliamentary debate proposed in Proksch and Slapin’s The Politics of Parliamentary Debate by combining a quantitative replication analysis of floor speeches with a qualitative explanation of the endogenous transformation of two institutions mediating elections and parliamentary speech in the Japanese Diet. Although the House of Representatives in Japan used a single non-transferable vote system with particularly strong personal vote-seeking incentives, our analysis based on a new dataset shows that, contrary to the theory’s predictions, the pattern of speech activities among party leaders and backbenchers is close to that of Germany, which has a party-centered proportional representation system. Such seemingly contradictory results, however, can be consistently understood within the scope of the theory if we consider the patterns of endogenous change in the parliamentary system and party organization. These results highlight the potential for further development of the theory.
Coauthored with Naoto Nonaka.
Party Politics. Published online before print.
[replication files]
3. “Pretending to Support? Duterte’s Popularity and Democratic Backsliding in the Philippines.”
The incumbent-led subversion of democracy represents the most prevalent form of democratic backsliding in recent decades. A central puzzle in this mode of backsliding is why these incumbents enjoy popular support despite their actions against democracy. We address this puzzle using the case of Philippine President Rodrigo Duterte. Although some Philippine analysts have speculated that his popularity was inflated due to social desirability bias (SDB) among survey respondents, there has been limited empirical examination. Our pre-registered list experiment surveys conducted in February/March 2021 detected SBD-induced overreporting at about 39 percentage points in face-to-face surveys and 28 percentage points in online surveys. We also found that the poor, Mindanaoans, and those who believed their neighbors supported Duterte, were more likely to respond according to SDB. These possibly counter-intuitive results should be interpreted with caution because the survey was conducted during the height of the COVID-19 lockdown, and the findings cannot necessarily be extrapolated to the other period of his presidency. Nevertheless, this study suggests that preference falsification could be an alternative explanation for the puzzle of popular incumbents in democratic backsliding.
Coauthored with Yuko Kasuya.
Journal of East Asian Studies 23 (3): 411–437. 2023.
[replication files]
4. “Gender Differences in Campaigning Under Alternative Voting Systems: Analysis of Election Manifestos.”
Growing evidence reveals that candidate issue engagement differs between men and women. However, recent research suggests that individual-level differences among candidates should be small under the strategic incentives inherent in single-member district elections that encourage party rather than personal vote seeking. We examine whether men and women candidates emphasize different issues in their electoral campaigns and if the magnitude of the gender gap varies under different electoral rules. Our analysis of 7497 Japanese election manifestos spanning more than 20 years, from 1986 to 2009, reveals significant gender differences in the issues candidates emphasize in their electoral campaigns, regardless of party affiliation or other attributes. Moreover, these differences remain salient after an extensive change from a multi-member district to single-member district electoral system.
Coauthored with Yoshikuni Ono.
Politics, Groups, and Identities 11 (5): 1203–1211. 2023.
[replication files]
5. “Voters’ Perceptions and Evaluations of Dynastic Politics in Japan.”
Political family dynasties are a staple part of Japanese politics. According to one study, Japan has the fourth highest number of dynastic politicians among democratic countries, after Thailand, the Philippines, and Iceland. As a result, many scholars have qualitatively studied how these political families are born and managed. In contrast to the abundance of qualitative research, however, very little quantitative research has focused on how Japanese voters view political dynasties. To understand this question, we conducted two nationwide surveys. Our major findings are that while the majority of respondents dislike dynastic candidates, they also value specific attributes of dynastic candidates, such as their political networks, their potential for ministerial appointments, and their ability to bring “pork projects” to their constituencies. These results serve as benchmark information on dynastic politics in Japan. They are also distinct from the findings of existing studies that Japanese voters are neutral about whether a candidate is from a dynastic family in voting decisions.
Coauthored with Yuko Kasuya and Yoshikuni Ono.
Asian Journal of Comparative Politics 8 (3): 671–688. 2023.
[replication files]
6. “Detecting Voter Understanding of Ideological Labels Using a Conjoint Experiment.”
Understanding voters’ conception of ideological labels is critical for political behavioral research. Conventional research designs have several limitations, such as endogeneity, insufficient responses to open-ended questions, and inseparability of composite treatment effects. To address these challenges, we propose a conjoint experiment to study the meanings ascribed to ideological labels in terms of policy positions. We also suggest using a mixture model approach to explore heterogeneity in voters’ understandings of ideological labels, as well as the average interpretation of labels. We applied these approaches to conceptions of left–right labels in Japan, where the primary issue of elite-level conflicts has been distinctive compared with other developed countries. We found that, on average, while Japanese voters understand policy-related meanings of “left” and “right,” they primarily associate these labels with security and nationalism, and, secondarily, with social issues; they do not associate these labels with economic issues. Voters’ understandings partly depend on their birth cohort, but observed patterns do not necessarily coincide with what many researchers would predict regarding generational differences in Japanese politics. Mixture model results suggest that some individuals tend to associate left–right labels with security and nationalism policies, while others link them to social policies. Over one-third of respondents seemed to barely understand the usage of left–right labels in policy positions. Our study improves upon existing methods for measuring voter understanding of ideological labels, and reconfirm the global diversity of meanings associated with left–right labels.
Coauthored with Reiko Arami and Masaki Taniguchi.
Political Behavior 45 (2): 635–657. 2023.
[replication files]
7. “Is the Justice Frame Effective in Mobilizing Support for Human Rights Violations? Evidence from the Philippines.”
Effective message framing motivates individuals to act for and defend human rights. What effective message framing motivates individuals to defend human rights? Recent experiment-based framing studies show that personal frames are more successful than informational or motivational frames in increasing the advocacy activities of human rights organizations. This study tested the justice frame using the Philippine case of extrajudicial executions. Employing internet-based survey experiments, we tested the effects of the justice frame on consensus and action mobilizations as well as the three frames mentioned above. Our results showed that combining justice and the personal frame is more effective. We also examined emotions inflicted by framing. Our results reveal an association between empathy and anger as a reaction that connects exposure to personal frame and mobilization.
Coauthored with Cleo Anne A. Calimbahin and Yuko Kasuya.
Asian Affairs: An American Review 50 (2): 77–94. 2023.
[replication files]
8. “What Do You Mean by ‘Democrat’ and ‘Republican’? Evidence from a Conjoint Experiment.”
To better understand the nature of American mass party identification, it is crucial to examine how voters understand terms used in measuring partisanship in public opinion surveys, such as “Democrat” and “Republican.” For this purpose, we conducted a conjoint experiment where respondents evaluated hypothetical profiles’ partisanship. We show that, in contrast to previous studies which rely on one theoretical perspective to understand mass partisanship, multiple theories are reflected in how voters understand labels representing their partisanship. Further, the results suggest that many Americans seem to equate party identification with vote choice, contrary to standard assumptions in American political behavior research that these are separate concepts.
Coauthored with Ikuma Ogura and Takeshi Iida.
International Journal of Public Opinion Research 34 (1): edab025. 2022.
[replication files]
9. “Are Voters Less Persuaded by Female Than by Male Politicians’ Statements? A Survey Experiment in Japan.”
We contribute to the demand-side theory of women’s underrepresentation in democracy, by which we mean that women are underrepresented in politics because voters prefer male politicians, by focussing on gender’s effect on the persuasive power of politicians’ policy statements. We conducted a survey experiment in Japan, where women are seriously underrepresented in politics and voter preferences for individual candidates matter with respect to women being elected. We showed participants fictitious local legislators’ policy statements and asked to what extent they supported those statements. We selected foreign and welfare issues on which voters have shared gendered stereotypes and randomly manipulated the legislators’ gender, ideological positions, and tone of their statements. We found no significant differences in approval for policy statements between female and male legislators. This result holds irrespective of legislators’ ideological positions and tone of the statements. We infer from the results of factual manipulation checks that such null results are attributable to the low rate of respondents who recognized the fictitious legislators’ gender. Although our study is preliminary, and its results are ambiguous, our experimental design helps scholars disentangle some mechanisms of voters’ underappreciation of female politicians.
Coauthored with Musashi Happo and Kaho Odaka.
Journal of Elections, Public Opinion and Parties 32 (2): 429–438. 2022.
[replication files]
10. “Why Do Citizens Prefer High-Skilled Immigrants to Low-Skilled Immigrants? Identifying Causal Mechanisms of Immigration Preferences with a Survey Experiment.”
Why do citizens prefer high-skilled immigrants to low-skilled immigrants? To understand the causal mechanism behind this tendency among citizens, we conducted a vignette survey experiment that enables us to clarify the role of multiple mediators. We specifically focused on three key factors that have been proposed in existing research as those that could lead citizens to welcome high-skilled immigrants: expectations of economic contribution, welfare contribution, and small crime potential. We found that the skill premium was fully eliminated when natives were informed that immigrants would be engaged in low-skill jobs in the host country, which underscores the essential role that post-migration work plays in the acceptance of immigrants by natives. Our findings provide suggestive evidence that natives welcome high-skilled immigrants simply because they expect economic benefits from high-skilled immigrants, not because they expect them to contribute to welfare or be less likely to commit crimes.
Coauthored with Akira Igarashi and Yoshikuni Ono.
Research and Politics 9 (2). 2022.
[replication files]
11. “Estimating Ideal Points of Newspapers from Editorial Texts.”
Although measuring the ideal points of news media is essential for testing political communication theories based on spatial theory, prior methods of estimating ideal points of media outlets have various shortcomings, including high cost in terms of time and human resources and low applicability to different countries. We propose that unsupervised machine learning techniques for text data, specifically the combination of a text scaling method and latent topic modeling, can be applied to estimate ideal points of media outlets. We applied our proposed methods to editorial texts of ten national and regional newspapers in Japan, where prior approaches are not applicable because newspapers have never officially endorsed particular parties or candidates, and because high-quality training data for supervised learning is not available. Our two studies, one of which analyzed editorials on a single typically ideological topic while the other investigated all editorials published by the target papers in one year, confirmed the popular view of Japanese newspapers’ ideological slant, which validates the effectiveness of our proposed approach. We also illustrate that our methods allow scholars to investigate which issues are closely related to the respective ideological positions of media outlets. Furthermore, we use the estimated ideal points of newspapers to show that Japanese people partially tend to read ideologically like-minded newspapers and follow such newspapers’ Twitter accounts even though their slant is not explicit.
Coauthored with Tomoki Kaneko and Taka-aki Asano.
International Journal of Press/Politics 26 (3): 719–742. 2021.
[replication files]
12. “Value Preferences and Structures among Japanese Voters and Political Candidates.”
This study investigated value preferences and structures among Japanese voters and political candidates. Voters were surveyed after the 2014 Lower House election, while political candidates were surveyed before the 2016 Upper House election. To measure their value preferences, respondents were asked to rank seven values: freedom, equality, economic stability, morality, self-reliance, social order, and patriotism. Statistical analyses reveal substantially different value priority between voters and candidates. Furthermore, although little distinctiveness in value preferences was observed among Japanese voters, the tendency and cohesion of value preferences among candidates varied across parties. More specifically, the four opposition parties that collaborated in the 2016 election and the Clean Government Party had similar value preferences, while the least cohesive party in terms of value preferences was the Liberal Democratic Party.
Japanese Political Science Review 4: 61–85. 2018.
[replication files]
13. “Can Reshuffles Improve Government Popularity? Evidence from a ‘Pooling the Polls’ Analysis.”
Scholars have recently argued that prime ministers reshuffle their cabinets strategically. Although some scholars assume that cabinet reshuffles help prime ministers increase their government’s popularity, this assumption has not been tested formally because of the endogeneity problem. In Japan, polling firms sometimes provide respondents with cues about a reshuffle when asking about cabinet approval following reshuffles, while others do not. I utilized this convention in the Japanese media to test the assumption that reshuffles increase cabinet approval ratings. Applying a dynamic linear model to pooled poll data from 2001 to 2015, I achieved high internal, external, and ecological validity. The analyses show that cues about reshuffles increase cabinet approval ratings by 2.4 percentage points on average, and the credible interval of the effect does not include zero. This result reinforces the findings of previous research on the theory of cabinet management.
Public Opinion Quarterly 82 (2): 322–342. 2018.
[replication files]
14. “Share the Name, Share the Vote: A Natural Experiment of Name Recognition.”
Do candidates garner more votes simply because their names are better recognized? To answer this question, we utilize elections to the Japanese House of Councillors as a natural experiment. Members are elected in national and local-level districts. To isolate the effect of name recognition on vote choice, we compare the vote shares of national district candidates in high name recognition prefectures—which we define as prefectures in which a national candidate shares the same surname as a local district candidate—and the other low name recognition prefectures. Our research design addresses internal and external validity problems from which previous studies suffer. We find that national candidates obtain 69% larger vote shares in high as opposed to low name recognition prefectures. This result holds when controlling for idiosyncratic characteristics of national candidates and prefecture-specific surname popularity.
Coauthored with Kentaro Fukumoto.
Journal of Politics 80 (2): 726–730. 2018.
[replication files]
15. “Heterogeneity in Voter Perceptions of Party Competition in Multidimensional Space: Evidence from Japan”
On the question of how voters perceive party positions, much of the existing literature has not adequately considered the case of a multidimensional policy space. Since the ideological cues related to each issue differ in a policy space of multiple dimensions, it is possible that each individual favors different dimensional cues. To test this hypothesis, this paper focuses on Japan’s 2012 Lower House election, which took place in a two-dimensional policy space. An analysis of how voters position the three major parties relative to each other reveals that there is actually heterogeneity in the voter perception of the policy space. Further, using a mixed logit model, we find that demographic factors and political attitudes explain this heterogeneity.
Coauthored with Masaki Taniguchi.
International Political Science Review 38 (5): 673–689. 2017.
[replication files]
16. 「Twitterデータによる日本の政治家・言論人・ 政党・メディアのイデオロギー位置の推定」
本稿は，Twitterのデータを用いて日本の政治家・言論人・政党・メディアのイデオロギー位置を推定する。政治家等のイデオロギー位置は様々な方法で推定されてきたが，Twitterのデータによる方法には，地方政治家や言論人のイデオロギー位置を推定できるなど多くのメリットがある。採用する統計モデルは，一般のTwitterユーザーが自分自身と似たイデオロギー位置の政治家等のアカウントをフォローすることを好むと仮定し，一般ユーザーが政治家等のアカウントをフォローしているか否かのデータを使って，両者の理想点を推定するものである。衆議院議員70人，参議院議員46人，政党など10団体・機関，新聞社6社，地方政治家・元政治家39人，言論人41人のイデオロギー位置を推定した。本稿の方法で推定された国会議員の位置を政治家調査データから推定した位置と比較すると高い相関を記録したことから，妥当な推定が行われたと評価できる。
『選挙研究』33 (1): 41–56．2017．
[オンライン付録] [再現用ファイル]
17. 「政治学における部分的観察可能性を伴うプロビットモデルとその拡張——有権者のイデオロギーのモデル化を例として」
部分的観察可能性を伴うプロビットモデルは，複数の潜在的な要因の複雑な組み合わせによって生じる離散的な結果を分析するための統計モデルである。政治学者は複数のアクターによる観察不可能な意思決定の相互作用や心理的要因が混在するサーベイの回答を分析することが多く，また，政治学では結果が離散変数として記録されることが多いため，このモデルは政治学の研究において有用である。本稿では，このモデルをその基礎にあるモデルとともに解説し，その政治学における適用や拡張を紹介する。さらに，イデオロギー的自己意識に関するサーベイの回答を分析するための応用モデルを構築する。イデオロギー的自己意識は回答者が自己を離散的なイデオロギー尺度上に位置づけることによって測定されるが，我々はこれを認識，強度，向きという3つの潜在的な要因に分解することができる。本稿の新しいモデルはマルコフ連鎖モンテカルロ法で推定できる。このモデルを日本の世論調査データに適用したところ，情報量規準によって新しいモデルが既存のモデルよりも優れていることが示された。また，既存のモデルからは導かれないいくつかの新しい結果を得た。
『行動計量学』43 (2): 113–128．2016．
[オンライン付録] [再現用ファイル]
18. “Voters’ Left–Right Perception of Parties in Contemporary Japan: Removing the Noise of Misunderstanding”
The prevailing theory states that either Japanese voters have stopped ideologically distinguishing parties or that the main political parties in Japan have become more centrist in recent years. These arguments are based on survey questions asking citizens to locate parties on an ideological scale. However, these questions may suffer from noise caused by respondents who misinterpret the question wording or answer the questions inappropriately to mask their misunderstanding of the terms “left” and “right.” To address this problem by extracting only the views of those who know the meaning of left–right terms, this article develops a mixture model. Applying the model to an opinion poll conducted after the 2012 Japanese general election, I confirm that those who comprehend the left–right terminology—slightly over half of all voters—largely perceived parties’ ideologies in the same way as experts. Additionally, I find that even these voters face difficulties in placing ambiguous or new parties on the political spectrum. This study has implications not only for understanding trends in Japanese political ideology but also for survey design and analysis of heterogeneous survey responses.
Japanese Journal of Political Science 16 (1): 114–137. 2015.
[supplementary information]
19. 「現代日本における争点態度のイデオロギー的一貫性と政治的洗練——Converseの呪縛を超えて」
Philip Converseの古典的研究以来，政治学では，イデオロギー的に考える能力をもつことは有権者の政治的洗練を意味するというのが支配的な見解である。しかし，イデオロギーのヒューリスティックとしての性質を考えると，それはむしろ政治的知識の不足した有権者のために必要なものである。本稿では，争点態度のイデオロギー的一貫性に焦点を当て，政治的知識の最も高い層と最も低い層は一貫性が低く，政治的知識を中程度にもつ層は一貫性が高いという仮説を立てた。加えて，この仮説は現代日本において支持されうることを論じた。争点態度のイデオロギー的一貫性を個人レベルで測ることのできる指標を利用して，2000年代以降に日本で行われた3つの世論調査のデータを分析した。分析結果は，仮説が予想するとおりに，有権者の政治的知識のレベルによって，争点態度のイデオロギー的一貫性に有意かつ実質的な曲線的関係があることを示した。この発見は政治的洗練の概念の再検討を促すものである。
『年報政治学』65 (1): 148–174．2014．
[web appendix] [R code]
【論文（査読なし）】
1. 「2021年総選挙における有権者の政策選好および争点の重要度と投票」
2022年2月に実施したオンライン調査のデータを用いて，2021年総選挙における政策選好および争点の重要性と投票政党の関係について考察した。政策選好に関しては，政治家調査のデータと本研究の調査のデータを統合して教師なし分類を適用し，対象者を9つのクラスターに分類した。その結果，自民党の立場を複数領域にまたがって支持する人はほとんどいないものの，同党は安保・原発に関して保守的な人々を中心に得票する一方，リベラル系野党は自身と近い政策選好をもつ人々の票すら固められなかったことが示唆された。争点の重要性に関しては，各回答者がどの争点を重視して候補者を評価するかを調べられるように工夫したコンジョイント実験を実施した。その結果，第1の分析で重要性が示唆された安保・原発争点だけでなく（それよりもむしろ）消費税や同性婚が重視されていることが明らかになった。
『選挙研究』38 (2): 48–62．2023．
[オンライン付録・再現用ファイル ]
2. 「イデオロギーに着目した日本の青年の政治的成熟度の検討」
本研究は，争点態度と党派性のイデオロギー的一貫性を有権者の能力の指標として，日本の青年が大人と比較してどの程度政治的に成熟しているのか，政治的成熟度が青年期にどのように向上するのかを調べる。我々は2017年に，一方は15〜22歳を対象とし，他方は有権者全体を対象とする，対となる2つのオンライン調査を実施し，左右イデオロギーに関連する様々な争点に関する意見を尋ねた。イデオロギー的に一貫した態度の割合をディリクレ回帰によって分析したところ，青年の争点態度は高齢者よりもイデオロギー的な一貫性が低いものの，15歳の時点でチャンスレベルよりも一貫性の高い態度をもっており，その一貫性は青年期を通じて上昇していた。我々は，本研究のデータと2017年総選挙の政治家調査データを組み合わせることで，党派性のイデオロギー的近接性についても調査した。標本選択順序プロビットモデルの推定結果によると，党派性をもっている人々に限れば，青年は大人と同程度にイデオロギー的に近い政党に対して党派性をもっていた。これらの結果は，日本における選挙権年齢の引き下げに関する議論に対して示唆を与える。
谷口将紀と共著．
『公共選択』75: 81–101．2021．
[オンライン付録] [再現用ファイル]
3. 「戦後日本人の憲法意識——世論調査集積法による分析」
本研究は，戦後に主要7調査機関が実施した憲法に関する世論調査の結果を包括的に分析することで，戦後日本人の憲法意識の変遷を追うことを目的とする。動的線形モデルを応用した世論調査集積法を用いることによって，質問内容やワーディングの違い，調査機関・調査方法ごとの傾向，標本誤差を考慮した上で，憲法改正に対する潜在的な賛成・反対率を推定できる。推定結果からは，有権者の認識において1950年代には憲法改正が全面改憲を意味したのに対して，1960〜80年代にかけて争点が9条改正に収斂していったこと，1990〜2000年代には9条以外の論点が明確に意識されるようになったこと，小泉政権後は焦点が再び9条問題に絞られつつあることが読み取れる。さらに，質問内容やワーディングに関する分析結果からは，一般的に9条の改正が2項の改正として有権者に認識されていることや，戦争を連想させることが9条改正の反対率を高めることなどが示唆される。
境家史郎と共著．
『年報政治学』71 (1): 34–57．2020．
[オンライン付録]
【その他の出版物】
1．「コロナ禍における世論に関する研究動向」
『統計』72 (11): 25–30．
[オンライン付録・再現用ファイル]
2．「『感情温度』が表すもの——東京大学×朝日新聞社の世論調査から」
吉田徹（編）『民意のはかり方——「世論調査×民主主義」を考える』法律文化社．2018．
[オンライン付録] [再現用ファイル]
訂正
本文の記述の誤りがありました。86頁に「『リベラル』と『イデオロギー』の意味をわからないと答えたのは，無回答の人を除いてそれぞれ27%，28%に上る」とありますが，これらの数字は分母に無回答の人を含んだものでした。無回答の人を分母から除くと，それぞれ30%，31%でした。お詫びして訂正いたします。