We compare the contents of Fox and MSNBC weekday evening telecasts using natural language analysis with Linguistic Inventory Word Count (LIWC) and sociopolitical dictionaries tapping into moral foundations, values, grievances, and personality. Across time, the two networks differed substantially across many constructs, particularly those from LIWC. The core of the difference between the networks was captured by a four-component measure which we labeled Personalizing versus Formal speech. Scores on this measure were particularly volatile during 2019 and 2020, a period which included Trump's first impeachment, the beginning of the COVID pandemic, and the 2020 presidential campaign. In comparison with prior presidential election years, only MSNBC showed a drop in positive emotions in 2020, while both networks increased in the use of communal and analytic language. Contrary to our expectations, the language style of the two networks did not demonstrate increasing divergence over time.

中文翻译：

---

换频道？Fox 和 MSNBC 在 2012 年、2016 年和 2020 年的比较

我们使用自然语言分析与语言库存字数 (LIWC) 和利用道德基础、价值观、不满和个性的社会政治词典来比较 Fox 和 MSNBC 平日晚间电视广播的内容。随着时间的推移，这两个网络在许多结构上存在很大差异，尤其是来自 LIWC 的那些。网络之间差异的核心是通过我们标记为个性化与正式的四分量测量来捕捉的演讲。该指标的得分在 2019 年和 2020 年期间尤其不稳定，这段时期包括特朗普的第一次弹劾、COVID 大流行的开始和 2020 年的总统竞选。与之前的总统选举年相比，只有 MSNBC 的积极情绪在 2020 年有所下降，而这两个网络在使用公共语言和分析语言方面都有所增加。与我们的预期相反，两个网络的语言风格并没有随着时间的推移表现出越来越大的分歧。