In this article, we examine the relationship between metrics documenting politics-related Twitter activity with election results and trends in opinion polls. Various studies have proposed the possibility of inferring public opinion based on digital trace data collected on Twitter and even the possibility to predict election results based on aggregates of mentions of political actors. Yet, a systematic attempt at a validation of Twitter as an indicator for political support is lacking. In this article, building on social science methodology, we test the validity of the relationship between various Twitter-based metrics of public attention toward politics with election results and opinion polls. All indicators tested in this article suggest caution in the attempt to infer public opinion or predict election results based on Twitter messages. In all tested metrics, indicators based on Twitter mentions of political parties differed strongly from parties’ results in elections or opinion polls. This leads us to question the power of Twitter to infer levels of political support of political actors. Instead, Twitter appears to promise insights into temporal dynamics of public attention toward politics.
Purpose – The steady increase of data on human behavior collected online holds significant research potential for social scientists. The purpose of this paper is to add a systematic discussion of different online services, their data generating processes, the offline phenomena connected to these data, and by demonstrating, in a proof of concept, a new approach for the detection of extraordinary offline phenomena by the analysis of online data.
Design/methodology/approach – To detect traces of extraordinary offline phenomena in online data, the paper determines the normal state of the respective communication environment by measuring the regular dynamics of specific variables in data documenting user behavior online. In its proof of concept, the paper does so by concentrating on the diversity of hashtags used on Twitter during a given time span. The paper then uses the seasonal trend decomposition procedure based on loess (STL) to determine large deviations between the state of the system as forecasted by the model and the empirical data. The paper takes these deviations as indicators for extraordinary events, which led users to deviate from their regular usage patterns.
Findings – The paper shows in the proof of concept that this method is able to detect deviations in the data and that these deviations are clearly linked to changes in user behavior triggered by offline events.
Originality/value – The paper adds to the literature on the link between online data and offline phenomena. The paper proposes a new theoretical approach to the empirical analysis of online data as indicators of offline phenomena. The paper will be of interest to social scientists and computer scientists working in the field.
Die Wahlkampf-Kampagne Barack Obamas von 2008 machte Twitter auch in Deutschland bekannt. Aber erst eine Kampagne gegen vermeintliche Internetzensur brachte die Veränderungen durch den Dienst auch hierzulande zum Vorschein: Politik wird zugänglicher und die Kommunikation schneller. Kann die Politik dem folgen?
As the microblogging service Twitter becomes an increasingly popular tool for politicians and general users to comment on and discuss politics, researchers increasingly turn to the relationship between tweets mentioning parties or candidates and their respective electoral fortunes. This paper offers a detailed analysis of Twitter messages posted during the run-up to the 2009 federal election in Germany and their relationship to the electoral fortunes of Germany’s parties and candidates. This analysis will focus on four metrics for measuring the attention on parties and candidates on Twitter and the relationship to their respective vote share. The metrics discussed here are: the total number of hashtags mentioning a given political party; the dynamics between explicitly positive or explicitly negative mentions of a given political party; the total number of hashtags mentioning one of the leading candidates, Angela Merkel (CDU) or Frank-Walter Steinmeier (SPD); and the total number of users who used hashtags mentioning a given party or candidate. The results will show that during the campaign of 2009 Twitter messages commenting on parties and candidates showed little, if any, systematic relationship with subsequent votes on election day. In the discussion of the results, I will raise a number of issues that researchers interested in predicting elections with Twitter will have to address to advance the state of the literature.