社會期許行為之試題反應模型

An Item Response Theory Model for Socially Desirable Behavior

冷承翰(Cheng-Han Leng)

指導教授 ： 姚開屏

國立臺灣大學/理學院/心理學研究所/碩士(2019年)

https://doi.org/10.6342/NTU201900056

摘要

本研究旨在建立一個新的試題反應理論模型(item response theory model)以考量個體在填答自陳量表(self-report assessment)時過度呈現(overreport)或低度呈現(underreport)真實自我的行為。此行為是根據欲測量特質之方向所產生，是一種在表達初始想法時會無意識欺騙自我，並下意識將選擇轉移至更符合讚許之選項的行為機制。此行為機制在本模型中以在多向度評等量尺模型(multidimensional rating scale model)內加入欺騙因子(deception term)，再另外乘上轉移因子(transfer term)的方式處理，其中轉移結因子以指標函數(indicator function)與轉移機率矩陣(transition probability matrix)所組成。本模型建立在貝式架構上(Bayesian framework)並以馬可夫鏈蒙地卡羅法(Markov chain Monte Carlo)估計參數。本模型之參數估計結果良好反映在一系列的模擬研究當中。此外，在實徵研究中，本模型被應用於分析由本研究所設計之網路實驗的資料，其中發現參與者傾向在記名時欺騙自我，亦在有記分板的狀況下轉移選擇。

關鍵字

社會期許 ； 試題反應理論 ； 多向度評等量尺模式

並列摘要

In this study, a new item response theory model is developed to account for the situation where respondents overreport or underreport their actual opinion corresponding to the positive or negative direction of an issue measured in a self-report assessment. Such behavior is conceptualized as a mechanism that not only deceives but transfers selections. In the proposed model, such mechanism is performed by incorporating a deception term into a multidimensional rating scale (MRS) model and then multiplying by a transfer term, which is performed by an indicator function and a transition matrix separately. The proposed model is presented in a Bayesian framework approximated by Markov Chain Monte Carlo (MCMC) algorithms. Through a series of simulations, the parameters under the proposed model are recovered well. The methodology is also implemented within an online experimental study to demonstrate its application.

並列關鍵字

socially desirable behavior ； item response theory model ； multidimensional rating scale model

參考文獻

Böckenholt, U. (2014). Modeling motivated misreports to sensitive survey questions. Psychometrika, 79, 515–537.

Bolt, D. M., & Lall, V. F. (2003). Estimation of compensatory and noncompensatory multidimensional item response models using markov chain monte carlo. Applied Psychological Measurement, 27, 395–414.

Brooks, S. P., & Gelman, A. (1998). General methods for monitoring convergence of iterative simulations. Journal of Computational and Graphical Statistics, 7, 434–455.

Brown, J. D. (1986). Evaluations of self and others: Self-enhancement biases in social judgments. Social Cognition, 4, 353–376.

Buss, A. H., & Perry, M. (1992). The aggression questionnaire. Journal of Personality and Social Psychology, 63, 452–459.