Thomas Justin, Al-Shehhi Aamna, Al-Ameri Marwa, Grey Ian
Zayed University, United Arab Emirates.
Khalifa University, United Arab Emirates.
Heliyon. 2019 Jul 26;5(7):e02087. doi: 10.1016/j.heliyon.2019.e02087. eCollection 2019 Jul.
Differences in self-concept have been observed across cultures. Participants from collectivist societies tend to describe themselves using social and relational attributes (mother, student, Arab) more frequently than their individualist counterparts, who tend to rely more heavily on personal attributes (fun, tall, beautiful). Much of this past research has relied on relatively small samples of college students, tasked with spontaneously reporting self-concepts in classroom settings. The present study re-examines these ideas using data extracted from Twitter, the popular social media platform. In analysis one, the Twitter biographies of individuals exclusively posting messages in English ( = 500) and those posting only in Arabic ( = 500) were content analyzed and quantified for differences in the frequency of personal versus social attribute use. Analysis two applied a bilingual word counting algorithm to the biographies of a larger sample of Twitter users ( = 242,162), exploring the relative frequency of social attributes, specifically familial roles (e.g. mother, father, daughter, son), across both English and Arabic users. In analysis one, the Twitter biographies of exclusive Arabic users contained significantly more social attributes than their English using counterparts. In analysis two, Arabic biographies contained significantly more familial references than their English language counterparts. These findings support the idea that cultural values may influence self-construal. Big data extracted from social media platforms appear to offer a useful means of exploring self-concept across cultures and languages.
不同文化间的自我概念差异已被观察到。集体主义社会的参与者倾向于比个人主义社会的参与者更频繁地使用社会和关系属性（母亲、学生、阿拉伯人）来描述自己，而个人主义社会的参与者则更倾向于依赖个人属性（有趣、高、漂亮）。过去的许多研究都依赖于相对较小的大学生样本，要求他们在课堂环境中自发报告自我概念。本研究使用从流行社交媒体平台推特提取的数据重新审视了这些观点。在分析一中，对仅用英语发帖的个人（=500）和仅用阿拉伯语发帖的个人（=500）的推特个人简介进行了内容分析，并对个人属性与社会属性使用频率的差异进行了量化。分析二则将双语单词计数算法应用于更大样本的推特用户（=242,162）的个人简介，探讨了英语和阿拉伯语用户中社会属性，特别是家庭角色（如母亲、父亲、女儿、儿子）的相对频率。在分析一中，仅使用阿拉伯语的用户的推特个人简介中包含的社会属性明显多于使用英语的用户。在分析二中，阿拉伯语个人简介中包含的家庭相关内容明显多于英语个人简介。这些发现支持了文化价值观可能影响自我建构的观点。从社交媒体平台提取的大数据似乎提供了一种跨文化和语言探索自我概念的有用方法。
