論文 - 星野　崇宏

Propensity Score Adjustment for Internet Panel Surveys of Voting Behavior: A Case in Japan

Tetsuro Kobayashi,Takahiro Hoshino

Japanese Journal of Electoral Studies 27 （ 2 ） 104 - 117 2011年

研究論文（学術雑誌）, 共著, 査読有り

Age and amount of exposure to a foreign language during childhood: Behavioral and ERP data on the semantic comprehension of spoken English by Japanese children

Shiro Ojima,Naoko Nakamura,Hiroko Matsuba-Kurita,Takahiro Hoshino,Hiroko Hagiwara

Neuroscience Research （Neuroscience Research）  70 （ 2 ） 197 - 205 2011年

研究論文（学術雑誌）, 共著, 査読有り,  ISSN  0168-0102

　概要を見る

Children's foreign-language (FL) learning is a matter of much social as well as scientific debate. Previous behavioral research indicates that starting language learning late in life can lead to problems in phonological processing. Inadequate phonological capacity may impede lexical learning and semantic processing (phonological bottleneck hypothesis). Using both behavioral and neuroimaging data, here we examine the effects of age of first exposure (AOFE) and total hours of exposure (HOE) to English, on 350 Japanese primary-school children's semantic processing of spoken English. Children's English proficiency scores and N400 event-related brain potentials (ERPs) were analyzed in multiple regression analyses. The results showed (1) that later, rather than earlier, AOFE led to higher English proficiency and larger N400 amplitudes, when HOE was controlled for; and (2) that longer HOE led to higher English proficiency and larger N400 amplitudes, whether AOFE was controlled for or not. These data highlight the important role of amount of exposure in FL learning, and cast doubt on the view that starting FL learning earlier always produces better results. © 2011 Elsevier Ireland Ltd and the Japan Neuroscience Society.

• Access to Document (DOI)

Statistical Significance Testing with Mahalanobis Distance for Thresholds Estimated from Constant Stimuli Method

Takehiro Nagai, Takahiro Hoshino, Keiji Uchikawa

Seeing and Perceiving （Seeing and Perceiving）  24 （ 2 ） 91 - 124 2011年

研究論文（学術雑誌）, 共著, 査読有り,  ISSN  1878-4755

　概要を見る

The t-test and the analysis of variance are commonly used as statistical significance testing methods. However, they cannot assess the significance of differences between thresholds within individual observers estimated from the constant stimuli method; these thresholds are not defined as averages of samples, but they are rather defined as functions of parameters of psychometric functions fitted to participants' responses. Moreover, the statistics necessary for these statistical testing methods cannot be derived. In this paper, we propose a new statistical testing method to assess the statistical significance of differences between thresholds estimated from the constant stimuli method. The new method can assess not only threshold differences but also main effects and interactions in multifactor experiments, exploiting the asymptotic normality of maximum likelihood estimators and the characteristics of multivariate normal distributions. This proposed method could be used in similar cases to the analysis of variance for thresholds estimated from the adjustment method and the staircase method. Finally, we present some data on simulations in which we tested assumptions, power and type I error of the proposed method. © 2011 BRILL.

• Access to Document (DOI)

Overview of the Japan children's study 2004-2009; Cohort study of early childhood development

Yamagata Z., Maeda T., Anme T., Sadato N., Koizumi H., Anji Y., Shiotani Y., Iwasaki M., Kutsuki A., Kuroki M., Ichikawa N., Morita T., Koike H., Morito Y., Cheng S., Ishida H., Yanaka H., Tanaka D., Namba K., Fukushi T., Toyoda H., Kimura-Ohba S., Sawada A., Wong K., Sakakihara Y., Kawaguchi H., Matsuishi T., Sogon S., Tomiwa K., Awaya T., Matuzawa S., Itakura S., Okada M., Komada Y., Yamamoto H., Yamakawa N., Bonno M., Momoi M., Yamagata T., Shiokawa H., Daisuke N Saito, Uchiyama H., Ozaki T., Ogura T., Ikeda H., Negayama K., Nakagawa K., Morimoto K., Kobayashi K., Koeda T., Tamaru T., Seki A., Terakawa S., Takeuchi A., Konishi Y., Sakura O., Kawai M., Egami S., Hoshino T., Yato Y.

Journal of Epidemiology （Journal of Epidemiology）  20 （ SUPPL.2 ） S397 - S403 2010年12月

ISSN  09175040

　概要を見る

Background: There are still a lot of unknown aspects about the childhood development of sociability which are based on neuroscientific basis. Purpose of the Japan Children's Study (JCS) was to verify the normal process of child development of sociability; the trajectory and factors related development of sociability, and to collect findings and integrate the knowledge to make the plan of long-term and large scale cohort study. Methods: A child cohort study underway in Japan since 2005. There are the cohort study including a infant cohort study at age of 4 months to 30 months and a preschool cohort study at age of 5 years old to 8 years old. Questionnaires, direct observation of children and cognitive testing were performed. Results: In infant cohort study, 465 infants were recruited at 4 months and 367 children were followed up to 30 months, follow up rate was 78.9% and in the preschool cohort study, total 192 children (112 at 2005 and 80 at 2007) at age of 5 years old and 169 followed up to 6 years (follow up rate was 88.0%), and 79 children were followed up to 8 years old (follow up rate was 70.5%) old. Several new measurements to evaluate child sociability were developed. Some factors related to development of child sociability were found for example the 'praise' was related to child sociability in cohort study based on neuroscience findings. Conclusions: Though the trajectory of child sociability development were not clarified, some significant factors related to development of sociability, and the basic findings to conduct a long-term and large scale cohort study were provided. © 2010 by the Japan Epidemiological Association.

• Access to Document (DOI)

Gender differences of children's social skills and parenting using Interaction Rating Scale (IRS)

Anme T., Shinohara R., Sugisawa Y., Tong L., Tanaka E., Watanabe T., Onda Y., Kawashima Y., Hirano M., Tomisaki E., Mochizuki H., Morita K., Gan-Yadam A., Yato Y., Yamakawa N., Yamagata Z., Koizumi H., Wong K., Anji Y., Ishida H., Iwasaki M., Kutsuki A., Kuroki M., Koike H., Saito D., Sawada A., Shiotani Y., Tanaka D., Cheng S., Toyoda H., Namba K., Fukushi T., Morita T., Yanaka H., Sakakihara Y., Morimoto K., Nakagawa K., Itakura S., Tomiwa K., Sogon S., Matsuishi T., Ogura T., Okada M., Ikeda H., Sadato N., Momoi M., Shiokawa H., Yamagata T., Maeda T., Ozaki T., Anme T., Hoshino T., Sakura O., Konishi Y., Kobayashi K., Koeda T., Tamaru T., Terakawa S., Seki A., Takeuchi A., Kawaguchi H., Egami S., Komada Y., Yamamoto H., Bonno M., Yamakawa N., Kawai M., Yato Y., Negayama K.

Procedia - Social and Behavioral Sciences （Procedia - Social and Behavioral Sciences）  2 （ 2 ） 260 - 268 2010年10月

ISSN  18770428

　概要を見る

Background: The purpose of this paper is to describe the gender differences of the Interaction Rating Scale (IRS) as an evidence-based practical index of children's social skills and parenting. Methods: The participants in our study, which was conducted as part of a Japan Science and Technology Agency (JST) cohort study project, were 370 dyads of children with their caregivers were followed up and surveyed at aged 18, 30, and 42 month. The participants completed the five minute interaction session and were observed using the IRS. Results: The results indicated the gender differences of IRS scores. Girls got rather high score at all ages. IRS can measure children's social skill development and parenting with high validity. Along with the qualitative differences of interaction between boys and girls, the IRS is effective in describing features of social skill development and parenting. Conclusion: The IRS provides further evidence of the fact that in order to study children's social skill development, it is important to evaluate various features of the caregiver-child interaction as a predictor of social skills. © 2010 Elsevier Ltd. All rights reserved.

• Access to Document (DOI)

Determining the direction of the path using a Bayesian semiparametric model

Kei Miyazaki, Takahiro Hoshino, Kazuo Shigemasu

Proceedings of 19th International Conference on Computational Statistics 19   1429 - 1436 2010年

研究論文（国際会議プロシーディングス）, 共著, 査読有り

Happiness is unique: A latent structure of emotion recognition traits revealed by statistical model comparison

Atsunobu Suzuki, Takahiro Hoshino, Kazuo Shigemasu

Personality and Individual Differences （Personality and Individual Differences）  48 （ 2 ） 196 - 201 2010年

研究論文（学術雑誌）, 共著, 査読有り,  ISSN  01918869

　概要を見る

Is an individual who is sensitive to the facial expressions of a given emotion also sensitive to those of other emotions? This study addressed this simple but fundamental question by examining the latent structure underlying the sensitivity scores for basic emotions expressed by faces (Suzuki, Hoshino, & Shigemasu, 2006a). Eight hundred and five participants took part in the study, and for each participant, the sensitivity scores for happiness, fear, anger, disgust, and sadness were calculated. Variants of a single- factor model were fit to the matrix of correlations among the five sensitivity scores. A solution for the best-fit model indicated an equal contribution from the single factor to the four negative emotions (path coefficient = 0.635) and a smaller contribution to happiness (0.183). Our results imply minimally over- lapped mechanisms underlying the recognition of positive and negative emotions. © 2009 Elsevier Ltd.

• Access to Document (DOI)

Web調査の偏りの補正: 行動経済学における調査研究への適用

星野　崇宏

関西大学ソシオネットワーク戦略研究機構 ワーキングペーパー 97   1 - 18 2010年

研究論文（学術雑誌）, 単著

調査不能がある場合の標本調査におけるセミパラメトリック推定と感度分析: 日本人の国民性調査データへの適用

星野　崇宏

統計数理 58 （ 1 ） 3 - 23 2010年

研究論文（学術雑誌）, 単著, 査読有り

Validity of caregivers' reports on head trauma due to falls in young children aged less than 2 years.

Fujiwara T, Nagase H, Okuyama M, Hoshino T, Aoki K, Nagashima T, Nakamura H

Clinical medicine insights. Pediatrics 4   11 - 8 2010年

潜在構造モデルによる複数カテゴリのセグメンテーション

小島 早都子, 田村 玄, 星野 崇宏

マーケティングサイエンス （日本マーケティング・サイエンス学会）  17 （ 1・2 ） 13 - 29 2009年03月

共著, 査読有り

就学前児の社会的スキルと広汎性発達障害(PDD)との関連

篠原　亮次, 星野　崇宏, 杉澤　悠圭, 童　連, 田中　笑子, 渡辺　多恵子, 恩田　陽子, 安梅　勅江

厚生の指標 12   20 - 25 2009年

研究論文（学術雑誌）, 単著, 査読有り

A New Concurrent Calibration Method for Nonequivalent Group Design under Nonrandom Assignment

Kei Miyazaki, Takahiro Hoshino, Shin-ichi Mayekawa, Kazuo Shigemasu

Psychometrika （Psychometrika）  74 （ 1 ） 1 - 19 2009年

共著, 査読有り,  ISSN  0033-3123

　概要を見る

This study proposes a new item parameter linking method for the common-item nonequivalent groups design in item response theory (IRT). Previous studies assumed that examinees are randomly assigned to either test form. However, examinees can frequently select their own test forms and tests often differ according to examinees' abilities. In such cases, concurrent calibration or multiple group IRT modeling without modeling test form selection behavior can yield severely biased results. We proposed a model wherein test form selection behavior depends on test scores and used a Monte Carlo expectation maximization (MCEM) algorithm. This method provided adequate estimates of testing parameters. © 2008 The Psychometric Society.

• Access to Document (DOI)

A Bayesian Semiparametric Item Response Model with Dirichlet Process Priors

Kei Miyazaki, Takahiro Hoshino

Psychometrika （Psychometrika）  74 （ 3 ） 375 - 393 2009年

共著, 査読有り,  ISSN  0033-3123

　概要を見る

In Item Response Theory (IRT), item characteristic curves (ICCs) are illustrated through logistic models or normal ogive models, and the probability that examinees give the correct answer is usually a monotonically increasing function of their ability parameters. However, since only limited patterns of shapes can be obtained from logistic models or normal ogive models, there is a possibility that the model applied does not fit the data. As a result, the existing method can be rejected because it cannot deal with various item response patterns. To overcome these problems, we propose a new semiparametric IRT model using a Dirichlet process mixture logistic distribution. Our method does not rely on assumptions but only requires that the ICCs be a monotonically nondecreasing function; that is, our method can deal with more types of item response patterns than the existing methods, such as the one-parameter normal ogive models or the two- or three-parameter logistic models. We conducted two simulation studies whose results indicate that the proposed method can express more patterns of shapes for ICCs and can estimate the ability parameters more accurately than the existing parametric and nonparametric methods. The proposed method has also been applied to Facial Expression Recognition data with noteworthy results. © 2009 The Psychometric Society.

• Access to Document (DOI)

ネット上の意見表明のバイアスとその補正について(<特集>情報の信頼性評価)

星野 崇宏

人工知能学会誌 （社団法人人工知能学会）  23 （ 6 ） 791 - 797 2008年11月

単著,  ISSN  9128085

就学前児の社会的スキル : コホート研究による因子構造の安定性と予測的妥当性の検討

高橋 雄介, 岡田 謙介, 星野 崇宏, 安梅 勅江

教育心理学研究 （Japanese Journal of Educational Psychology）  56 （ 1 ） 81 - 92 2008年03月

共著, 査読有り,  ISSN  215015

　概要を見る

Social skills are positive abilities that help children to interact with others in various situations. Many previous studies have found that failure in acquiring social skills or atypical development of social skills is predictive of later problem behavior, social maladjustment, and low academic achievement. Currently, however, few scales exist to measure the social skills of Japanese preschoolers. In Study 1, a large-scale cohort study was used to develop a scale that enables an appropriate measure of individual differences in preschoolers' social skills. The obtained 3-factor structure was consistent with published studies done in Western countries, and the scale was found to have excellent internal consistency and temporal stability. In Study 2, the relation between the newly developed scale, problem behavior, and autistic tendencies was analyzed. The results confirmed that the present scale demonstrates both concurrent and predictive validity. It is possible that early intervention based on screening with this scale might assist children to develop soundly.

• Access to Document (DOI)

ブランドイメージに関する広告政策を策定するための階層ベイズ的な選択モデルとその応用

星野 崇宏

マーケティングサイエンス （日本マーケティング・サイエンス学会）  15 （ 1・2 ） 27 - 44 2008年03月

単著, 査読有り

組織風土と不祥事に関する実証分析

星野　崇宏

一橋経済学 2 （ 2 ） 53 - 73 2008年

単著

Standard errors of estimated latent variables with estimated structural parameters

Takahiro Hoshino, Kazuo Shigemasu

Applied Psychological Measurement （Applied Psychological Measurement）  32 （ 2 ） 181 - 189 2008年

共著, 査読有り,  ISSN  01466216

　概要を見る

The authors propose a concise formula to evaluate the standard error of the estimated latent variable score when the true values of the structural parameters are not known and must be estimated. The formula can be applied to factor scores in factor analysis or ability parameters in item response theory, without bootstrap or Markov chain Monte Carlo methods. The proposed method is valid when both the number of items and the number of subjects are large. However, simulation results suggest that the proposed method provides valid estimates for standard errors of latent abilities even when the numbers of items and subjects are not large. © 2008 Sage Publications.

• Access to Document (DOI)

A Bayesian Propensity Score Adjustment for Latent Variable Modeling and MCMC algorithm

Takahiro Hoshino

Computational Statistics & Data Analysis （Computational Statistics and Data Analysis）  52 （ 3 ） 1413 - 1429 2008年

単著, 査読有り,  ISSN  01679473

　概要を見る

The estimation of the differences among groups in observational studies is frequently inaccurate owing to a bias caused by differences in the distributions of covariates. In order to estimate the average treatment effects when the treatment variable is binary, Rosenbaum and Rubin [1983. The central role of the propensity score in observational studies for causal effects. Biometrika 70, 41-55] proposed an adjustment method for pre-treatment variables using propensity scores. Imbens [2000. The role of the propensity score in estimating dose-response functions. Biometrika 87, 706-710] extended the propensity score methodology for estimation of average treatment effects with multivalued treatments. However, these studies focused only on estimating the marginal mean structure. In many substantive sciences such as the biological and social sciences, a general estimation method is required to deal with more complex analyses other than regression, such as testing group differences on latent variables. For latent variable models, the EM algorithm or the traditional Monte Carlo methods are necessary. However, in propensity score adjustment, these methods cannot be used because the full distribution is not specified. In this paper, we propose a quasi-Bayesian estimation method for general parametric models that integrate out the distributions of covariates using propensity scores. Although the proposed Bayes estimates are shown to be consistent, they can be calculated by existing Markov chain Monte Carlo methods such as Gibbs sampler. The proposed method is useful to estimate parameters in latent variable models, while the previous methods were unable to provide valid estimates for complex models such as latent variable models. We also illustrated the procedure using the data obtained from the US National Longitudinal Survey of Children and Youth (NLSY1979-2002) for estimating the effect of maternal smoking during pregnancy on the development of the child's cognitive functioning. © 2007 Elsevier B.V. All rights reserved.

• Access to Document (DOI)