SEARCH

Search Details

BUNJI Kyosuke
Graduate School of Business Administration / Division of Business Administration
Associate Professor

Researcher basic information

■ Research Keyword
  • Multiple-alternative Forced-choice Format
  • Paired Comparison Format
  • Computer Based Testing
  • Response Time
  • Reaction Time
  • Item Response Theory
  • Bayesian statistics
  • Adaptive testing
  • Rater Effect
■ Research Areas
  • Humanities & social sciences / Educational psychology / Educational measurement
  • Informatics / Statistical science / Cognitive psychometrics
■ Committee History
  • Apr. 2025 - Present, 一般社団法人 社会調査協会, 編集委員
  • Jan. 2025 - Present, The Japanese Association of Educational Psychology, Editorial Board, Japanese Journal of Educational Psychology
  • Apr. 2023 - Present, Common Achievement Test Organization (CATO), Test Reliability and Validity Review Committee
  • Jul. 2021 - Present, 日本行動計量学会, 運営委員会
  • Aug. 2025 - Aug. 2025, 日本テスト学会, 第23回大会実行委員
  • Mar. 2025 - Mar. 2025, 日本行動計量学会, 第27回春の合宿セミナー実行委員
  • Nov. 2024 - Nov. 2024, 日本行動計量学会, 2024秋の行動計量セミナー実行委員長
  • Sep. 2024 - Sep. 2024, 日本行動計量学会, 第52回大会実行委員
  • Mar. 2024 - Mar. 2024, 日本行動計量学会, 第26回春の合宿セミナー実行委員
  • Nov. 2023 - Nov. 2023, 日本行動計量学会, 2023秋の行動計量セミナー実行委員
  • Mar. 2023 - Mar. 2023, 日本行動計量学会, 第25回春の合宿セミナー実行委員
  • Nov. 2022 - Nov. 2022, 日本行動計量学会, 2022秋の行動計量セミナー実行委員
  • Mar. 2022 - Mar. 2022, 日本行動計量学会, 第24回春の合宿セミナー実行委員

Research activity information

■ Award
  • Sep. 2024 The Japanese Association of Educational Psychology, Kido Award, Development of a short form of the Children’se Emotion Vocabulary Scale, using item response theory analysis
    Mai Hamana, Kyosuke Bunji
    Official journal

  • May 2020 日本計算機統計学会, 奨励賞
    分寺杏介
    本学会の欧文誌に公刊された Bunji & Okada (2019) は,反応時間もモデルに組み込んだ項目反応理論の優れた論文であり,本学会シンポジウムでも優れた口頭発表をしている (分寺・岡田,2017,2018)。その他,Behaviorresearch methods,日本テスト学会誌や,行動計量学の諸誌に,テストデータの計算機統計学の優れた論文を公刊しており,今後の研究が奨励されます。 Bunji, K., & Okada, K. (2019). Item response and response time model for personality assessment via linear ballistic accumulation. Japanese Journal of Statistics and Data Science, 2, 263–297. 分寺杏介,岡田謙介 (2018) Thurstonian IRT におけるパラメータの不変性について.日本計算機統計学会第32回シンポジ ウム. 分寺杏介,岡田議介 (2017) 反応時を利用する新たな項目反応モデ ルの提案 ― Linear Ballistic Accumulation を用いた効率的な推定法 ―.日本計算機統計学会第31回シンポジウム.
    Japan society

  • Dec. 2017 日本テスト学会, 日本テスト学会大会発表賞, 出題領域と解答時間を考慮した項目選択アルゴリズムの性能評価
    登藤直弥, 分寺杏介, 尾崎幸謙
    Japan society

  • Nov. 2017 日本計算機統計学会, 学生研究発表賞, 反応時間を利用する新たな項目反応モデルの提案 -Linear Ballistic Accumulation を用いた効率的な推定法-
    分寺杏介
    Japan society

■ Paper
  • Yoshito Tan, Keishi Nomura, Kyosuke Bunji, Kensuke Okada
    This study developed a public database on the degree of perceived social desirability for personality trait words. Perceived social desirability scores of 420 trait words were obtained from 500 respondents. Further, the procedures for trait word selection and data collection were described in detail. High correlations were observed between the perceived social desirability scores in this study and those in four previous studies. Finally, the usefulness of the database was discussed.
    Japan Society of Personality Psychology, Jul. 2025, The Japanese Journal of Personality, 34(1) (1), 88 - 90, Japanese, Domestic magazine
    [Refereed]
    Scientific journal

  • Ayako Baba, Kyosuke Bunji, Aya Fukushima
    This study defines “stumbling in response (hereinafter called "stumbling”)” as the difficulty in giving a valid response about their condition to a certain situation, psychological scale, or item. This study explored the presence, situation, and clinical psychologists’ support for clients’ stumbling in clinical situations from the perspective of clinical psychologists. An open-ended web-based survey of clinical psychologists showed that 99 out of 103 valid respondents noticed stumbling. Qualitative analysis revealed various situations causing stumbling, the diversity of clients’ expressions, clinical psychologists’ support, and different characteristics of stumbling in the collected responses. Quantitative analysis of co-occurrence frequencies and rates showed that the situation was related to stumbling, and that stumbling was related to support. These results are expected to facilitate the development of appropriate methods for stumbling prevention and support.
    The Japanese Psychological Association, Jan. 2025, 心理学研究, 96(1) (1), 48 - 54, Japanese, Domestic magazine
    [Refereed]
    Scientific journal

  • Kyosuke Bunji, Kensuke Okada
    A fundamental assumption in item response theory (IRT) models for multiple-alternative forced-choice (MAFC) personality measurement is that the utility of each statement (choice option) remains independent and invariant, irrespective of factors like the context effect. However, investigations on the confounding impact of the context effect on MAFC personality measurements are scarce. This study examined the impact of context effect on the utility and estimation of the parameters of each statement of the MAFC personality measurement. Thereon, a parameter estimation was conducted using the Thurstonian IRT model, assuming the presence or absence of context effects. Parameter estimates and Bayesian posterior distributions were compared from the perspective of the correlation of trait scores, an information criterion, explained variance, 95\% credible interval, the Bayes factor, and the overlap rate of the posterior distributions. The study's results suggest that the context effect can be considered small enough to be practically negligible.
    Sep. 2024, PsyArXiV (preprint), English

  • Kyosuke Bunji
    This paper selectively reviews the outline and current developments on theoretical aspects of computer-based testing (CBT) in various fields. Six major research topics were introduced in detail: score comparability between PBT and CBT, adaptive testing, innovative item types, frauds and their countermeasures in online testing, utilizing log data, and reasonable accommodation. In addition, following the previous predictions on the development of CBTs and the literature review, the future direction of research on CBTs was discussed from three perspectives: validity, test anxiety, and automation.
    The Japan Association for Research on Testing, Jun. 2023, Japanese Journal for Research on Testing, 19(1) (1), 191 - 225, Japanese, Domestic magazine
    [Refereed]
    Scientific journal

  • Ayako Baba, Kyosuke Bunji
    May 2023, JMIR Mental Health, 10, Article e42420, English, International magazine
    [Refereed]
    Scientific journal

  • Mai Hamana, Kyosuke Bunji
    Acquisition of emotion words plays an important role in the socialization or the development of children's emotional experience. This study aimed to develop emotion words vocabulary scales for preschooler and young elementary school children by applying the model of Item Response Theory (IRT). In study 1, data was collected from 1136 mothers with 3 to 9-year-old children. Based on the data, short form of the emotion vocabulary scale was constructed. Study 2 and 3 reports the pattern of correlation relationships of the emotion vocabulary scale with several relevant variables. The results showed that emotion vocabulary was associated with language ability and social emotional competence in children; thus the validity evidence of the scale was confirmed.
    The Japanese Association of Educational Psychology, Mar. 2023, The Japanese Journal of Educational Psychology, 71(1) (1), 51 - 61, Japanese, Domestic magazine
    [Refereed]
    Scientific journal

  • Seito Nonoda, Kyosuke Bunji, Kensuke Okada
    Paired-comparison format questionnaires that measures two latent traits suffer from the problem of unidentifiability in the Thurstonian item response model. This study proposes a new method that enables us to identify this model. Specifically, we pro- pose to use every item twice in the questionnaire. In simulation studies, our proposed method recovered the respondent parameters at least as well as the control method, which required analysts to fix some of the parameters to their true values. In the real data analysis, the proposed method was applied to data obtained by two types of psychological scales, one of which contained only positively keyed items, whereas the other contained both positively and negatively keyed items. Item parameters and trait scores estimated using the former data resulted in large posterior standard deviations. In contrast, those estimated with the latter data exhibited better estimation proper- ties. Although it generally does not allow for precise measurement, we show that the proposed method is applicable to empirical analysis. It achieved progress, in that, it negates the need for the unrealistic assumption of fixing parameters to their true values.
    The Behaviormetric Society, Sep. 2021, The Japanese journal of behaviormetrics, 48(2) (2), 53 - 68, Japanese, Domestic magazine
    [Refereed]
    Scientific journal

  • Kyosuke Bunji, Kensuke Okada
    Mar. 2021, Multivariate Behavioral Research, 57(4) (4), 658 - 678, English, International magazine
    [Refereed]
    Scientific journal

  • Bayesian statistical modeling for modern personality measurement
    Kyosuke Bunji, Kensuke Okada
    コンピュータ上で行う心理特性の測定には,回答に要した時間も収集できることや,回答バイアスの影響を受けにくい比較判断型項目を自然に実装ができることといった利点がある。従来の紙の調査票をそのままコンピュータに載せるのではなく,こうした特徴を活用して妥当性の高い測定を行えると望ましい。本稿ではこれを実現するための,ベイズ統計的アプローチに基づく項目反応理論モデルの拡張による私たちの研究を紹介する。
    一般社団法人社会調査協会, Sep. 2020, Advances in Social Research, 25, 22 - 30, Japanese, Domestic magazine
    [Invited]
    Scientific journal

  • Kyosuke Bunji, Kensuke Okada
    Jun. 2020, Behavior Research Methods, 52(3) (3), 1091 - 1107, English, International magazine
    [Refereed]
    Scientific journal

  • Kensuke Okada, Kyosuke Bunji
    Much scholarly attention is focused on psychological measurement through the paired-comparison format, which is considered to be tolerant to systematic response bias. The Thurstonian D-diffusion item response theory model was recently proposed to incorporate response-time information in this context. Because reliability is a fundamental measurement property, this study used the above model to conduct a preliminary investigation into the extent of the reliability increase achieved when incorporating response-time information into paired-comparison psychological measurement. Under some realistic conditions, our simulation revealed a practically relevant (but not very large) degree of increase. The same type of increase was also found during our analysis of a real psychological dataset containing measurements for the Big Five traits. As such, this study produced evidence supporting the collection and utilization of response time when conducting paired-comparison psychological measurement.
    The Behaviormetric Society, Apr. 2020, Behaviormetrika, 48(1) (1), 169 - 177, English, International magazine
    [Refereed]
    Scientific journal

  • Kyosuke Bunji
    The University of Tokyo, Mar. 2020, 博士学位論文(東京大学教育学研究科), English
    Doctoral thesis

  • Kyosuke Bunji
    This study compared several item selection algorithms in a situation where a time limit exists in terms of estimation accuracy and item bank security. Simulation results show the following: (1) adjusting item information using the expected response time will improve both the estimation accuracy and the item bank security; (2) when item discrimination and expected response time are positively correlated, the mean numbers of the answered items decrease particularly in the algorithms which determine the next item according to the information; (3) under the condition in (2), algorithms that suppress the inflation of overlap rate demonstrate better estimation accuracy and; (4) the root-mean-square error and test information show slightly different results between algorithms.
    The Japan Association for Research on Testing, Jun. 2019, Japanese Journal for Research on Testing, 15(1) (1), 1 - 20, Japanese, Domestic magazine
    [Refereed]
    Scientific journal

  • Kyosuke Bunji, Kensuke Okada
    Apr. 2019, Japanese Journal of Statistics and Data Science, 2(1) (1), 263 - 297, English, International magazine
    [Refereed]
    Scientific journal

  • Teiko Arai, Kyosuke Bunji, Naoya Todo, Noriko H. Arai, Takuya Matsuzaki
    Jul. 2018, Proceedings of the 40th Annual Cognitive Science Society Meeting (CogSci 2018), 100 - 105, English
    [Refereed]
    International conference proceedings

  • Arai Teiko, Bunji Kyosuke, Ishihara Yuki, Matsuzaki Takuya, Kageura Kyo
    This paper analyses linguistic features of primary school textbooks for the penultimate and final grades (grades five and six) and junior high school textbooks for the first and second grades (grades one and two), in order to reveal factors affecting the so-called "primary-secondary" learning gap. We chose science textbooks and analysed their complexities from the points of view of lexical and syntactic characteristics. The analysis revealed that there is a clear gap in syntactic complexities of sentences between primary and secondary school textbooks. The analysis showed that the factors constitute the gap between primary and secondary school textbooks.
    The Mathematical Linguistic Society of Japan, Sep. 2017, Mathematical Linguistics, 31(2) (2), 144 - 159, Japanese, Domestic magazine
    [Refereed]
    Scientific journal

  • Reading skill test to diagnose basic language skills in comparison to machines
    Noriko H. Arai, Naoya Todo, Teiko Arai, Kyosuke Bunji, Shingo Sugawara, Miwa Inuzuka, Takuya Matsuzaki, Koken Ozaki
    CORE A
    2017, Proceedings of the 39th Annual Cognitive Science Society Meeting (CogSci 2017), 1556 - 1561, English, International magazine
    [Refereed]
    International conference proceedings

  • Kyosuke Bunji
    This study proposes a graded response model including a new parameter, which represents rater effects affecting inter-item variance. The validity of this model was shown by parameter restoration simulation, in which artificial item response data were generated and calibrated under 12 conditions each. Another simulation study, which compares estimation accuracy of θ in the presented model and original graded response model, was conducted. Results indicated the proposed model can remove the rater effect. Using a rating experiment that induced the halo effect, it was shown that the new parameter wir can be regarded as a halo effect.
    The Behaviormetric Society, Sep. 2016, The Japanese journal of behaviormetrics, 43(2) (2), 181 - 195, Japanese, Domestic magazine
    [Refereed]
    Scientific journal

■ MISC
  • Book Review: Matrix-Based Introduction to Multivariate Data Analysis: Second edition
    Kyosuke Bunji
    Jun. 2021, Japanese Journal for Research on Testing, 17(1) (1), 128 - 129, Japanese
    Book review

  • Emotion Vocabulary Development in Early Childhood : Based on Item Response Theory
    Mai Hamana, Kyosuke Bunji
    最終版はこちら↓ 浜名真以, 分寺杏介 (2023). 項目反応理論を用いた幼児および小学校低学年用短縮版感情語彙尺度の開発. 教育心理学研究. 71(1). 51-61. https://doi.org/10.5926/jjep.71.51
    明治安田こころの健康財団, 2018, Research-aid paper of the Yasuda Life Welfare Foundation, 54, 105 - 114, Japanese, Domestic magazine
    Others

  • Yoshiki Ishikawa, Hirotake Ito, Aya Uemura, Shin Tabata, Risako Toyama, Kyosuke Bunji, Takahiro Hoshino, Ryo Matsuoka, Kazuhiro Yamaguchi
    本稿では、就学期の子どもの認知能力を計測する目的で行われている「学力テスト」のあり方を展望し、今後の教育政策をエビデンスに基づくものに転換していくために、どのような学力テストのあり方が望ましいかを論じる。まずは国内外で実施された学力テストの動向を振り返り、その後平成27年から開始された「埼玉県・学力学習状況調査」を1つのモデルケースとして取り上げる。本稿の後半では、平成27年から開始された「埼玉県・学力学習状況調査」のデータの簡単な記述統計と分析を紹介するとともに、これと同時に行われた埼玉県戸田市での教員調査の結果も踏まえ、今後の改善点についても論じる。
    独立行政法人経済産業研究所, Mar. 2017, RIETI Policy Discussion Paper, 17-P-010, Japanese
    Technical report

■ Lectures, oral presentations, etc.
  • Desirability-matched Thurstonian IRT scale construction leveraging sentiment analysis
    Kensuke Okada, Yoshito Tan, Keishi Nomura, Kotaro Tsuchida, Kyosuke Bunji
    IMPS 2025, Jul. 2025, English, Psychometric Society, International conference
    Oral presentation

  • Statistical models and scaling techniques for forced-choice psychological measurement
    Kensuke Okada, Kyosuke Bunji, Keishi Nomura, Yoshito Tan, Hiroshi Shimizu, Koji E Kosugi
    The 52nd Annual Meeting of the Behaviormetric Society, Sep. 2024, Japanese, 2024_BMS_bunji_okada.pdf, No password, The Behaviormetric Society, Osaka University of Economics, Osaka, Japan, Japan, Domestic conference
    Public symposium

  • A method for presenting multiple-alternative personality measurement items to reduce cognitive load
    Kyosuke Bunji, Takeshi Sugiyama
    The 52nd Annual Meeting of the Behaviormetric Society, Sep. 2024, Japanese, 2024_BMS_bunji_sugiyama.pdf, No password, The Behaviormetric Society, Osaka University of Economics, Osaka, Japan, Japan, Domestic conference
    Oral presentation

  • Capturing the "difficult-to-answer" feeling towards psychological scales: A quantitative approach assessing individual subjectivity
    Kyosuke Bunji, Ayako Baba, Aya Fukushima
    The 88th Annual Convention of the Japanese Psychological Association, Sep. 2024, Japanese, 2024_JPA_Bunji_Baba_Fukushima.pdf, No password, The Japanese Psychological Association, Kumamoto-jo Hall, Kumamoto, Japan, Japan, Domestic conference
    Poster presentation

  • An investigation of the context effect in the multidimensional forced-choice personality measurement
    Kyosuke Bunji, Kensuke Okada
    IMPS 2024, Jul. 2024, English, IMPS_2024_Tuesday_RB101_Bunji.pdf, No password, Psychometric Society, Prague University of Economics and Business (Prague, Czech), Czech Republic, International conference
    Oral presentation

  • Past, present, and future of computer based testing
    Kyosuke Bunji
    17th Commemorative Lecture of the Japan Association for Research on Testing, Mar. 2024, Japanese, テスト学会セミナー2024_bunji.pdf, Published, No password, The Japan Association for Research on Testing, Online, Japan, Domestic conference
    [Invited]
    Public discourse

  • Do psychological scales measure clients' minds?: Focusing on "stumbling" in responses and support from psychologists
    Ayako Baba, Kyosuke Bunji, Aya Fukushima
    The 87th Annual Convention of the Japanese Psychological Association, Sep. 2023, Japanese, 日本心理学会2023_2D-033-PD.pdf, Published, No password, The Japanese Psychological Association, Kobe International Conference Center, Hyogo, Japan, Japan, Domestic conference
    Poster presentation

  • An examination of order effects in paired comparison psychological measurement
    Kyosuke Bunji
    The 51st Annual Meeting of the Behaviormetric Society, Aug. 2023, Japanese, 行動計量2023_bunji.pdf, Published, No password, The Behaviormetric Society, Aoyama Gakuin University, Tokyo, Japan, Japan, Domestic conference
    Oral presentation

  • An examination of the stability of paired comparison psychological measurement under different response methods
    Kyosuke Bunji, Kensuke Okada
    The 37th Annual Meeting of Japanese Society of Computational Statistics, Jun. 2023, Japanese, Japanese Society of Computational Statistics, Tokachi Plaza, Hokkaido, Japan, Japan, Domestic conference
    Oral presentation

  • A Bayesian statistical approach to examining the comparability of CBT and PBT scores
    Kyosuke Bunji
    The 50th Annual Meeting of the Behaviormetric Society, Aug. 2022, Japanese, the Behaviormetric Society, Okinawa Shichoson Jichikaikan, Okinawa, Japan, Japan, Domestic conference
    Oral presentation

  • An examination of the influence of anchor item selection methods in equating with common item designs
    Kyosuke Bunji
    The 19th Annual Meeting of the Japan Association for Research on Testing, Sep. 2021, Japanese, Japan Association for Research on Testing, Online, Japan, Domestic conference
    Oral presentation

  • Detection of mental health problem using annual student health survey data
    Ayako Baba, Kyosuke Bunji
    The 85th Annual Convention of the Japanese Psychological Association, Sep. 2021, Japanese, 日本心理学会, オンライン, Japan, Domestic conference
    Poster presentation

  • An extension of the LBA-IRT model to multidimensional multiple-alternative forced-choice personality measurement with response times
    Kyosuke Bunji, Kensuke Okada
    MathPsych/ICCM 2020, Jul. 2020, English, Society for Mathematical Psychology, オンライン, International conference
    Oral presentation

  • An LBA-IRT model for Likert data
    Kyosuke Bunji, Kensuke Okada
    The 33rd Symposium of Japanese Society of Computational Statistics, Nov. 2019, Japanese, 日本計算機統計学会, 青山学院大学青山キャンパス(東京都渋谷区), Japan, Domestic conference
    Oral presentation

  • Development of a Children’s Emotion Vocabulary Scale using item response theory
    Mai Hamana, Kyosuke Bunji
    The 83rd Annual Convention of the Japanese Psychological Association, Sep. 2019, Japanese, Japanese Psychological Association, Ritsumeikan University, Osaka, Japan, Domestic conference
    Poster presentation

  • An item response model for multidimensional multiple-alternative forced-choice data using response time
    Kyosuke Bunji, Kensuke Okada
    The 47th Annual Meeting of the Behaviormetric Society, Sep. 2019, Japanese, The Behaviormetric Society, Osaka University, Osaka, Japan, Japan, Domestic conference
    Oral presentation

  • Cognitive process model for ordered categorical response and response time data for questionnaires
    Kyosuke Bunji, Kensuke Okada
    MathPsych/ICCM 2019, Jul. 2019, English, Le Centre Sheraton (Montreal, QC), Canada, International conference
    Oral presentation

  • Observed-score reliability and its approximate index in paired-comparison Thurstonian IRT
    Kensuke Okada, Kyosuke Bunji
    IMPS 2019, Jul. 2019, English, Psychometric Society, International conference
    Oral presentation

  • On parameter invariance in Thurstonian IRT model
    Kyosuke Bunji, Kensuke Okada
    The 32th Symposium of Japanese Society of Computational Statistics, Nov. 2018, Japanese, Japanese Society of Computational Statistics, Shiga University, Shiga, Japan, Japan, Domestic conference
    Oral presentation

  • Performance evaluation of item selection algorithms considering item domain and answer time
    Naoya Todo, Kyosuke Bunji, Koken Ozaki
    The 46th Annual Meeting of the Behaviormetric Society, Sep. 2018, Japanese, The Behaviormetric Society, Keio University, Tokyo, Japan, Japan, Domestic conference
    Oral presentation

  • An extension of linear ballistic accumulator model for rank data
    Kyosuek Bunji
    The 46th Annual Meeting of the Behaviormetric Society, Sep. 2018, Japanese, The Behaviormetric Society, Keio University, Tokyo, Japan, Japan, Domestic conference
    Oral presentation

  • Modeling Forced-choice Version of Questionnaires on the Big Five Factors for Reducing Response Biases
    Kyosuke Bunji, Kensuke Okada
    MathPsych/ICCM 2018, Jul. 2018, English, Society for Mathematical Psychology, University of Wisconsin–Madison (Madison, Wisconsin), United States, International conference
    Oral presentation

  • Extending the diffusion-IRT model to forced-choice response time data
    Kyosuke Bunji, Kensuke Okada
    IMPS 2018, Jul. 2018, English, Psychometric Society, Columbia University (New York City, USA), United States, International conference
    Oral presentation

  • A new item response model using response time: efficient estimation method via linear ballistic accumulation
    Kyosuek Bunji, Kensuke Okada
    The 31st Symposium of Japanese Society of Computational Statistics, Nov. 2017, Japanese, Japanese Society of Computational Statistics, Wakayama Medical University, Wakayama, Japan, Japan, Domestic conference
    Oral presentation

  • A model for analyzing paired comparison data using response time
    Kyosuke Bunji
    The 45th Annual Meeting of the Behaviormetric Society, Aug. 2017, Japanese, The Behaviormetric Society, University of Shizuoka, Shizuoka, Japan, Japan, Domestic conference
    Oral presentation

  • Performance evaluation of item selection algorithms considering item domain and answer time
    Naoya Todo, Kyosuke Bunji, Koken Ozaki
    The 15th Annual Meeting of the Japan Association for Research on Testing, Aug. 2017, Japanese, The Japan Association for Research on Testing, Tohoku University, Miyagi, Japan, Japan, Domestic conference
    Oral presentation

  • An exploratory study of the relationship between reading comprehension and reading, life, and study habits: using the Reading Skills Test
    Kyosuke Bunji, Noriko Arai
    The 15th Annual Meeting of the Japan Association for Research on Testing, Aug. 2017, Japanese, The Japan Association for Research on Testing, Tohoku Univeristy, Miyagi, Japan, Japan, Domestic conference
    Oral presentation

  • New IRT Model Incorporating Response Time via Linear Ballistic Accumulation
    Kyosuke Bunji, Kensuke Okada
    MathPsych/ICCM 2017, Jul. 2017, English, The Society for Mathematical Psychology, University of Warwick (Coventry, United Kingdom), United Kingdom, International conference
    Poster presentation

  • A quantitative analysis of text reading difficulty: A case study of elementary and junior high school science textbooks
    Teiko Arai, Kyosuke Bunji, Takuya Matsuzaki, Kyo Kageura
    IPSJ SIG Computers and the Humanities, May 2017, Japanese
    Poster presentation

  • Reading Skills Test (RST): Its educational measurement properties and implications for educational practice
    Noriko Arai, Shingo Sugawara, Koken Ozaki, Miwa Inuzuka, Teiko Arai, Kyosuke Bunji, Hiroyuki Noguchi, Naoya Todo
    The 59th Annual Meeting of the Japanese Association for Educational Psychology, Oct. 2017, Japanese, The Japanese Association for Educational Psychology, Nagoya University, Aichi, Japan, Japan, Domestic conference
    Public symposium

  • Factor structure in personality trait tests conducted in several countries
    Shino Sako, Yutaro Sakamoto, Shiho Imashiro, Kyosuke Bunji
    The 14th Annual Meeting of the Japan Association for Research on Testing, Sep. 2016, Japanese, The Japan Association for Research on Testing, The University of Electro-Communications, Tokyo, Japan, Japan, Domestic conference
    Oral presentation

  • Proposal of a marginal model for linear combination of multiple traits on a multidimensional item response theory model
    Kyosuke Bunji, Takahiro Hoshino
    The 44th Annual Meeting of the Behaviormetric Society, Sep. 2016, Japanese, The Behaviormetric Society, Sapporo Gakuin University, Hokkaido, Japan, Japan, Domestic conference
    Oral presentation

  • An item response model that incorporates rater bias affecting variance among raters and examinees
    Kyosuke Bunji
    The 43rd Annual Meeting of the Behaviormetric Society, Sep. 2015, Japanese, The Behaviormetric Society, Tokyo Metropolitan University, Tokyo, Japan, Japan, Domestic conference
    Poster presentation

  • Impact of the question wording on social studies test: question setting considering the measuring intent of the examiner
    Kazuhiro Yasunaga, Kyosuke Bunji, Hiroyuki Ito, Yuji Mitobe
    The 57th Annual Meeting of the Japanese Association for Educational Psychology, Aug. 2015, Japanese, The Japanese Association for Educational Psychology, Japan, Domestic conference
    Poster presentation

■ Affiliated Academic Society
  • Japanese Society of Computational Statistics

  • The Japan Association for Research on Testing

  • The Behaviormetric Society

  • The Japanese Psychological Association

  • The Japanese Association of Education Psychology

■ Research Themes
  • Uncovering Common Determinants of Creativity Across Individuals, Teams, and Organizations Using AI-Based Reinforcement Learning Models
    原田 勉, 滝口 哲也, 分寺 杏介, 陳 金輝, 庭本 佳子, 原 泰史
    Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research, Grant-in-Aid for Scientific Research (B), Kobe University, Apr. 2025 - Mar. 2030, Coinvestigator
    Competitive research funding

  • Development of a foundation of comparative psychological measurement and its application
    Kensuke Okada, Izumi Matsuda, Tetsuya Kawamoto, Kyosuke Bunji, Keishi Nomura
    Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research, Grant-in-Aid for Scientific Research (A), The University of Tokyo, 01 Apr. 2025 - 31 Mar. 2029, Coinvestigator

  • Development of a Machine Learning Approach for Estimating Psychological Traits Using Diverse Paradata
    分寺 杏介
    Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research, Grant-in-Aid for Early-Career Scientists, Kobe University, Apr. 2024 - Mar. 2028, Principal investigator

  • 岡田 謙介, 分寺 杏介
    科学技術振興機構, AIP加速研究, Apr. 2025 - Mar. 2026, Coinvestigator
    本研究では、社会的望ましさバイアスの影響を抑えながら多数の対象者から効果的に非認知特性の測定を行うことができる、比較型測定に基づくテストの社会実装を行います。そのための技術開発と実証研究を進め、企業や教育の現場で実用化のためのケーススタディを行います。非認知特性の信頼性の高い測定を通じて、個人の特性に基づいた最適な人材配置やチーム構築、個別最適な指導を志向し、組織や社会の幸福度向上を目指します。

  • Kyosuke Bunji, Ayako Baba
    The Telecommunication Advancement Foundation, Grants for research, Kobe University, Apr. 2024 - Mar. 2026, Principal investigator
    Competitive research funding

  • 情報セキュリティ教育ツールによる情報セキュリティ意識変容の調査に関する学術研究
    分寺 杏介
    LRM株式会社, 共同研究, Dec. 2024 - Nov. 2025, Principal investigator
    Industry academia cooperation

  • Development of an adaptive multiple forced-choice psychological measurement using response time
    分寺 杏介
    Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research, Grant-in-Aid for Research Activity Start-up, Kobe University, 31 Aug. 2022 - 31 Mar. 2024, Principal investigator
    Competitive research funding

  • Modeling the rater effect by considering cognitive processes within individuals
    Kyosuke Bunji
    Japan Society for the Promotion of Science, Grants-in-Aid for Scientific Research, Grant-in-Aid for JSPS Fellows, The University of Tokyo, 26 Apr. 2017 - 31 Mar. 2020, Principal investigator
    本研究の目的は,多様な評定者効果を統一的に表現する数理モデルを提案し,その性能を評価すること,また提案モデルを使用した実際の分析例を提示することであった. 本年度は前年度に引き続き,Linear Ballistic Accumulationモデル(LBA)に基づく項目反応モデルについての考察および,Thurstoneの比較判断の法則に基づくモデルの拡張に関する研究を中心に行った.本年度の研究としては,第一にLBA IRTモデルを多次元多肢選択項目に拡張するためのモデルの拡張を行った(業績4,論文投稿準備中).Thurstoneの比較判断の法則を援用することにより,各選択肢の潜在的な効用を考慮したモデリングを行うことができると考えられる.また,LBAモデルをベースにすることで回答時間データを併せて用いることが可能となり,より豊かな情報を引き出すことが可能になると考えられる.将来的には,選択肢間での潜在的効用を促進・抑制しあう構造を取り入れることで,文脈効果を考慮したモデリングが可能になると期待される.第二にLBA IRTモデルをリッカート尺度などの順序データに適用するための拡張の研究を行った(業績5,6).通常のLBA IRTモデルは二択の項目にしか適用できないため,一問あたりの情報量が限られてしまう.リッカート尺度は心理学に限らず様々な分野で広く用いられている手法であるため,本モデルの提案により,より一般的な場面においてもLBA IRTを適用することが可能となり,また二択の場合と比べてより情報量の多い分析が可能になると考えられる.
    Competitive research funding

■ Social Contribution Activities
■ Academic Contribution Activities
■ Others
  • 東京大学FFP(フューチャー・ファカルティ・プログラム)第9期修了
    Apr. 2017 - Jul. 2017

TOP