Moving beyond the dominant bag-of-words approach to sentiment analysis we introduce an alternative procedure based on distributed word embeddings. The strength of word embeddings is the ability to capture similarities in word meaning. We use word embeddings as part of a supervised machine learning procedure which estimates levels of negativity in parliamentary speeches. The procedure's accuracy is evaluated with crowdcoded training sentences; its external validity through a study of patterns of negativity in Austrian parliamentary speeches. The results show the potential of the word embeddings approach for sentiment analysis in the social sciences.

Sentiment is important in studies of news values, public opinion, negative campaigning or political polarization and an explosive expansion of digital textual data and fast progress in automated text analysis provide vast opportunities for innovative social science research. Unfortunately, tools currently available for automated sentiment analysis are mostly restricted to English texts and require considerable contextual adaption to produce valid results. We present a procedure for collecting fine-grained sentiment scores through crowdcoding to build a negative sentiment dictionary in a language and for a domain of choice. The dictionary enables the analysis of large text corpora that resource-intensive hand-coding struggles to cope with. We calculate the tonality of sentences from dictionary words and we validate these estimates with results from manual coding. The results show that the crowdbased dictionary provides efficient and valid measurement of sentiment. Empirical examples illustrate its use by analyzing the tonality of party statements and media reports.

Research on negative campaigning has grown rapidly in the past decades. This article reviews the literature dealing with this campaign strategy. It discusses its definition and measurement and stresses the mismatch between the academic literature and general perceptions. It then reviews why parties and candidates choose to 'go negative' with a particular focus on the rationales for negative campaigning under multi-party competition. The manuscript further discusses the literature on electoral effects and broader societal consequences of negative campaigning and emphasizes issues related to data collection and research designs. The conclusion summarizes the state of the art and outlines avenues for future research.

Parties try to shape media coverage in ways that are favorable to them, but what determines whether media outlets pick up and report on party messages? Based on content analyses of 1,496 party press releases and 6,512 media reports from the 2013 Austrian parliamentary election campaign, we show that media coverage of individual party messages is influenced not just by news factors, but also by partisan bias. The media are therefore more likely to report on messages from parties their readers favor. Importantly, this effect is greater rather than weaker when these messages have high news value. These findings have important implications for understanding the media's role in elections and representative democracies in general.

Parties and politicians want their messages to generate media coverage and thereby reach voters. This article examines how attributes related to content and sender affect whether party messages are likely to get media attention. Based on content analyses of 1,613 party press releases and 6,512 media reports in a parliamentary, multiparty context, we suggest that party messages are more likely to make it into the news if they address concerns that are already important to the media or other parties. Discussing these issues may particularly help opposition parties and lower-profile politicians get media attention. These results confirm the importance of agenda setting and gatekeeping, shed light on the potential success of party strategies, and have implications for political fairness and representation.