Anonymity is one of the defining characteristics of computer-mediated communication (CMC), and numerous studies have demonstrated that anonymity influences communication behavior. Furthermore, individuals are motivated to express emotions as part of maintaining interpersonal relationships, and this holds true on social networking services (SNS) as well. However, few studies have directly examined the relationship between anonymity and emotional expression on SNS platforms. Therefore, this study aims to explore whether emotional expression in posts on SNS "X" differs depending on the degree of anonymity associated with the user's account. Data were collected in October 2024 using the Twitter Developer API Ver. 2.2. Based on the emotion word list developed by Miura (2015), 600 posts containing positive emotion words and 600 posts containing negative emotion words were extracted, yielding a total of 1,200 posts. From these, 395 posts were selected for analysis after excluding those from corporate accounts, bots, and reposts. To control for the degree of anonymity, we adopted the typology of personal information on SNS proposed by Sato (2012). Personal information consisted of three types: demographic information, traceable information, and private information. Posts were classified into eight groups based on whether or not these types of information were disclosed. Accounts that disclosed all types of information were considered to have low anonymity, while accounts that disclosed none were considered to have high anonymity. The results of sentiment analysis indicated that posts from accounts with lower anonymity tended to express more positive sentiment. Conversely, posts from accounts with higher anonymity tended to express more negative sentiment. These findings suggest that the degree of account anonymity may be associated with the positivity or negativity of sentiment expression in user-generated posts on social networking sites.