行動計量学
Online ISSN : 1880-4705
Print ISSN : 0385-5481
ISSN-L : 0385-5481
45 巻 , 1 号
選択された号の論文の6件中1~6を表示しています
特集 調査法研究の新展開2
  • 三浦 麻子, 小林 哲郎
    原稿種別: 特集 調査法研究の新展開2
    2018 年 45 巻 1 号 p. 1-11
    発行日: 2018年
    公開日: 2018/11/03
    ジャーナル フリー

    This study investigated the influence of satisficing on response behavior in online surveys. We compared online response data to psychological scales and logical thinking tasks conducted by an online survey company and a crowd sourcing service. In previous studies, satisficing was found to be more likely to occur among online survey monitors than among crowd sourcing service contributors. Results of the present study replicated it and showed that satisficing in terms of inattentively reading items significantly damages the integrity of psychological scales. On the other hand, it was also found that the influence of satisficing in terms of inattentively reading instructions carefully can be reduced by raising respondents' awareness. Those conducting online surveys should discuss taking active measures to minimize satisficing.

  • 北條 大樹, 岡田 謙介
    原稿種別: 特集 調査法研究の新展開2
    2018 年 45 巻 1 号 p. 13-25
    発行日: 2018年
    公開日: 2018/11/03
    ジャーナル フリー

    This study conducts a data-driven classification of the response styles for the 2,131 respondents of the SHARE (Survey of Health, Ageing and Retirement in Europe) survey. In the standard Likert scale measurement, item responses reflect not only the latent traits of respondents but also their response style biases which are irrelevant for the purpose of the original measurement. The anchoring vignettes is an effective method to measure and correct for such biases. In this study, we first modeled the anchoring vignettes variables in the SHARE dataset using the Bayesian multidimensional item response model. Then, we classified the estimated individual response style parameters using the divisive analysis clustering. As a result, seven different clusters of response styles were obtained. While some of them correspond to the well-documented response styles, many of the clusters of respondents exhibit unique response styles which are both interpretable and relevant. Thus, bottom-up classification approach of response styles would undertake a key role in revealing the empirical analysis of item response behavior.

原著
  • 橋本 翔, 田中 一晶, 片平 建史, 長田 典子
    原稿種別: 特集 調査法研究の新展開2
    2018 年 45 巻 1 号 p. 27-38
    発行日: 2018年
    公開日: 2018/11/03
    ジャーナル フリー

    In this study, we proposed a new analytical method for three-mode data to explain the individual differences that exist independently in changes of stimuli. Three-mode data, such as semantic differential data, is widely used in such fields as psychology, marketing studies, or Kansei engineering. We assumed that the response of an individual to a stimulus is determined by component scores consisting of the sum of the scores of the stimulus and tendencies of the person. We introduced a new component model that consists of a loading matrix, stimuli scores, and individual tendencies, and we minimized it with the alternating least squares algorithm. The results of simulation studies and the analysis of two data sets illustrate the validities and utilities of the method.

  • 財津 亘, 金 明哲
    原稿種別: 特集 調査法研究の新展開2
    2018 年 45 巻 1 号 p. 39-47
    発行日: 2018年
    公開日: 2018/11/03
    ジャーナル フリー

    This study examined the accuracy for author identification by text mining. We conducted 16 analyses (four writing styles × four multivariate analyses) across texts of 100 Bloggers, written by approximately 1,000 characters. Specifically, we conducted (1) principal components analysis, (2) correspondence analysis, (3) multi-dimensional scaling, and (4) hierarchical cluster analysis on each writing style: (1) rate of usage of non-independent words, (2) bigram of parts-of-speech, (3) bigram of postpositional particles, and (4) positioning of commas. We obtained high accuracy: 100% on sensitivity and 95.1% on specificity. Furthermore, the results showed no effects of age and gender against accuracy for author identification.

  • 末廣 徹, 武田 浩一, 神津 多可思, 竹村 敏彦
    原稿種別: 特集 調査法研究の新展開2
    2018 年 45 巻 1 号 p. 49-58
    発行日: 2018年
    公開日: 2018/11/03
    ジャーナル フリー

    In this paper, we analyze the current inflation perceptions, the formation mechanism of the 5-year inflation expectations, and their heterogeneity by gender and age with Structural Equation Modeling (SEM) based on “Survey on consciousness of individual investors” conducted in February 2016. We find that 1) if the current anxiety, which is a latent variable related to anxiety about the current Japanese economy and the state of household, is higher, the current inflation perceptions become higher; 2) the current inflation perceptions also affect the 5- year inflation expectations; and 3) “future anxiety” doesn't directly affect inflation expectations. 4) If the “future anxiety” become stronger, the inflation expectations are boosted through “current anxiety” and inflation perceptions. 5) Young people and elderly people have higher inflation perceptions. This U-shaped tendency could be explained by their strong “current anxiety.” 6) Inflation expectations tend to be higher in the 50s and 60s regardless of their strong future and current anxiety. 7) Male's inflation perceptions tend to be low, but another analysis shows that there is no gender difference in inflation expectations.

資料
  • 永野 駿太, 木本 茉莉奈, 小杉 考司, 小野 史典
    原稿種別: 特集 調査法研究の新展開2
    2018 年 45 巻 1 号 p. 59-66
    発行日: 2018年
    公開日: 2018/11/03
    ジャーナル フリー

    This study examined theeffect of knowing the answer as a factor affecting the illusion of transparency. Previous studies showed that people have the illusion of transparency, or a tendency to believe that their internal states are more discernible to others than they actually are. Meanwhile, previous studies showed that people have a tendency to overestimate the extent of easiness of presented tasks when they know the answers. We divided 70 subjects into 3 groups; actors (n=12), observers being taught the answer (n=29) and observers not being taught the answer (n=29). Actors drank five cups of soft drink, including a vinegar-mixed one while pretending to be impassive in front of a video camera, and estimated the number of correct identifiers for the vinegar-mixed one out of 10 observers. Observers being taught the answer estimated the same thing as actors after watching each actors' video. Observers not being taught the answer tried to identify the number of the vinegar-mixed one correctly after watching each actors' video. As compared to the actual number of correct identifiers, actors and observers being taught the answer estimated more correct identifiers. The result implied that knowing the answer is one of the factors affecting the illusion of transparency. Furthermore, the result showed that the magnitude of illusion caused by knowing the answers is approximately 10 percent and the magnitude of illusion caused by subjective experience of this experimental task (“drink vinegar-mixed soft drink”) is approximately 11 percent.

feedback
Top