In this article, we review existing research on the complementarity of social media data and survey data for the study of public opinion. We start by situating our review in the extensive literature (N = 187) about the uses, challenges, and frameworks related to the use of social media for studying public opinion. Based on 187 relevant articles (141 empirical and 46 theoretical) - we identify within the 141 empircal ones six main research approaches concerning the complementarity of both data sources. Results show that the biggest share of the research has focused on how social media can be used to confirm survey findings, especially for election predictions. The main contribution of our review is to detail and classify other growing complementarity approaches, such as comparing both data sources on a given phenomenon, using survey measures as a proxy in social media research, enriching surveys with SMD, recruiting individuals on social media to conduct a second survey phase, and generating new insight on “old” or “under-investigated” topics or theories using SMD. We discuss the advantages and disadvantages associated with each of these approaches in relation to four main research purposes, namely the improvement of validity, sustainability, reliability, and interpretability. We conclude by discussing some limitations of our study and highlighting future paths for research.
17 March 2022 in International Journal of Information Management Data Insights
Automated text analysis methods have made it possible to classify large corpora of text by measures such as frames and tonality, with a growing popularity in social, political and psychological science. These methods often demand a training dataset of sufficient size to generate accurate models that can be applied to unseen texts. In practice, however, there are no clear recommendations about how big the training samples should be. This issue becomes especially acute when dealing with texts skewed toward categories and when researchers cannot afford large samples of annotated texts. Leveraging on the case of support for democracy, we provide a guide to help researchers navigate decisions when producing measures of tonality and frames from a small sample of annotated social media posts. We find that supervised machine learning algorithms outperform dictionaries for tonality classification tasks. However, custom dictionaries are useful complements of these algorithms when identifying latent democracy dimensions in social media messages, especially as the method of elaborating these dictionaries is guided by word embedding techniques and human validation. Therefore, we provide easily implementable recommendations to increase estimation accuracy under non-optimal condition.
Social media have increasingly been used by political bodies and experts to disseminate health information to the public. However, we still know little about how the communication of these actors on social media is received by other users and how it reflects trends in public trust. We examined social media dynamics in the communication of information by major actors (n = 188) involved in COVID-19 online discussions in Switzerland. These actors are scientists (experts), policymakers (government officials, cantonal executives, and other parties), and representatives of mass media. We found little correlation between Twitter features (other users' engagement and negativity in other users' replies) and the level of public trust found in representative opinion surveys. We used topic modelling in combination with correspondence analysis, and including additional variables for actor types and the period of the public debate further enabled us to detect salient episodes related to the pandemic on social media. In particular, we found that differing roles were played by the (health) experts and political authorities in terms of both topics and influence on the specific timing of the pandemic. The results of this study provide helpful conclusions for communication among political authorities, health experts, and the public.
16 September 2022 in Swiss Political Science Review
Notwithstanding growing research on how using social media for political campaigning impacts politicians' chances of winning votes, we still have limited knowledge about whether and how the use of social media and online styles of communication affect political success over successive legislatures. We address this deficit by analyzing a panel dataset about the Twitter activity of politicians who have had a parliamentary mandate at least once. We first demonstrate that politicians' interaction with specific online audiences (e.g. in terms of replies and mentions) is still evolving, thus pointing to possible strategic adaptations of politicians' communication as social media are mastered. Then, we show that Twitter‐based activity moderately impacts politicians' political success, both in terms of political ranking and media coverage. This success, however, strongly depends on the style of political communication and on the legislature under scrutiny.
28 November 2022 in Information Processing & Management