Gender bias in political discourse is a significant problem on today's social media. Previous studies found that the gender of politicians indeed influences the content directed towards them by the general public. However, these works are particularly focused on the global north, which represents individualistic culture. Furthermore, they did not address whether there is gender bias even within the interaction between popular journalists and politicians in the global south. These understudied journalist-politician interactions are important (more so in collectivistic cultures like the global south) as they can significantly affect public sentiment and help set gender-biased social norms. In this work, using large-scale data from Indian Twitter we address this research gap. We curated a gender-balanced set of 100 most-followed Indian journalists on Twitter and 100 most-followed politicians. Then we collected 21,188 unique tweets posted by these journalists that mentioned these politicians. Our analysis revealed that there is a significant gender bias -- the frequency with which journalists mention male politicians vs. how frequently they mention female politicians is statistically significantly different ($p<<0.05$). In fact, median tweets from female journalists mentioning female politicians received ten times fewer likes than median tweets from female journalists mentioning male politicians. However, when we analyzed tweet content, our emotion score analysis and topic modeling analysis did not reveal any significant gender-based difference within the journalists' tweets towards politicians. Finally, we found a potential reason for the significant gender bias: the number of popular male Indian politicians is almost twice as large as the number of popular female Indian politicians, which might have resulted in the observed bias. We conclude by discussing the implications of this work.

翻译：政治话语中的性别偏见是当今社交媒体上的一个显著问题。先前研究发现，政治人物的性别确实会影响公众对其言论的内容。然而，这些研究主要聚焦于代表个人主义文化的全球北方，且未探讨在全球南方的流行记者与政治人物互动中是否存在性别偏见。这类未被充分研究的记者-政治人物互动（在集体主义文化如全球南方中更为重要）可能显著影响公众情绪，并助长性别偏见的社会规范。本研究利用印度推特的大规模数据填补了这一空白。我们选取了推特上关注度最高的100名印度记者（男女各半）和100名政治人物（男女各半），收集了这些记者提及上述政治人物的21,188条原创推文。分析表明存在显著的性别偏见：记者提及男性和女性政治人物的频率存在统计显著性差异（$p<<0.05$）。事实上，女记者提及女性政治人物的推文中位数获得的点赞量，比提及男性政治人物的推文中位数少十倍。然而，在推文内容分析中，情绪得分分析和主题建模分析均未发现记者对政治人物的推文存在显著性别差异。我们最终发现了潜在原因：印度流行男性政治人物数量是流行女性政治人物数量的近两倍，这可能导致观察到的偏差。最后，我们讨论了本研究的启示。