AUTHOR=Zhu Gancheng , Zhou Yuci , Zhou Fengfeng , Wu Min , Zhan Xiangping , Si Yingdong , Wang Peng , Wang Jun 

TITLE=Proactive Personality Measurement Using Item Response Theory and Social Media Text Mining

JOURNAL=Frontiers in Psychology

VOLUME=Volume 12 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2021.705005

DOI=10.3389/fpsyg.2021.705005

ISSN=1664-1078

ABSTRACT=This study intends to propose a novel method of assessing proactive personality by combining text mining and item response theory (IRT) to measure proactive personality more efficiently. We got free texts (essay question text dataset and micro-blog text dataset) and item response data of 901 college students. To enhance validity and reliability, 3 methods were employed in the study. Method 1 uses item response data to develop a proactive personality evaluation model based on IRT; Method 2 uses free texts to establish a proactive personality evaluation model based on text mining; Method 3 put the text mining results as the prior information for the IRT estimation and build a proactive personality evaluation model combining text mining and IRT. Finally, we evaluate the 3 methods using the confusion matrix method. The major result shows that: 1) Proposed combined method using essay question text, micro-blog text, and IRT’s estimated ability parameters performs the highest accuracy of 0.849; 2) Proposed combined method using essay question text and IRT’s estimated ability parameters has the highest sensitivity of 0.821; 3) The text classification method using essay question text data works best, reaching 0.959 in specificity measurement; 4) If the model is considered comprehensively, the combined method using essay question text + micro-blog text + IRT estimated ability parameters achieves 0.844 in F1 score. Thus, we can get the conclusion the novel combination method is significantly better than the traditional method.