Social media sentiment as an information signal in economy and society

Project background and research questions

Social media (e.g., Twitter) have become ubiquitous in today’s society and play an increasingly important role for both established firms and new ventures (Moser, Tumasjan & Cable, 2020).

Building on our prior work on Twitter sentiment as a signal for forecasting election results (Tumasjan, Sprenger, Sandner & Welpe, 2010; 2011), predicting stock prices (Sprenger, Tumasjan, Sandner & Welpe, 2014), and identifying firms’ stock-relevant news events (Sprenger, Sandner, Tumasjan & Welpe, 2014), in two current research projects we focus on two distinct but related research questions in the entrepreneurship context:

  1. Twitter sentiment and venture capital-based startups’ valuations and success:
    How does Twitter sentiment about new technologies and trends (e.g., blockchain, 5G, 3-D printing) influence venture capitalists’ (VC’s) valuations of new ventures whose business model is based on these technologies and trends, and, ultimately, these ventures’ actual success (i.e., acquisition or IPO)?
  2. Twitter-based personality analysis and venture capitalists’ investment success:
    Can Twitter sentiment-based personality analysis (Big Five personality model) of venture capitalists’ (VC’s) personality predict their investment success?


In both projects, we collect and analyze large data sets and use natural language (NLP) processing techniques for analyzing Twitter sentiment. In project 1, we analyze the sentiment about 37 technologies and trends in more than 412,000 Tweets from 2008 to 2017. In project 2, we extract personality measures from the Twitter timelines of 911 VCs (on average 785 Tweets and 5,000 words for each VC) to predict their investment success in more than 8,200 investments.

Project publications and conference presentations

Tumasjan, A., Braun, R., & Stolz, B. (2021). Twitter sentiment as a weak signal in venture capital financingJournal of Business Venturing, 36(2), 106062.

Spreng, S., Braun, R., & Tumasjan, A. (2020). Investor personality and success in venture capital. 80th Academy of Management Annual Meeting, Vancouver, British Columbia, Canada, USA.

Tumasjan, A., Braun, R., & Stolz, B. (2019). Give me a signal: How sentiment and patents drive venture capital valuations. 79th Academy of Management Annual Meeting, Boston, Massachusetts, USA.

Stolz, B., Braun, R. & Tumasjan, A. (2018) Venture capital investments in technology-based startups: Hype or sustainable success. G-Forum: 22nd Annual Interdisciplinary Conference on Entrepreneurship, Innovation and SMEs, Stuttgart, Germany.


Moser, K. J., Tumasjan, A., & Cable, D. (2020). Don't be so emotional: How social media communication affects potential applicants' engagement. Proceedings of the Twenty-Eighth European Conference on Information Systems (ECIS2020).

Sprenger, T. O., Tumasjan, A., Sandner, P. G., & Welpe, I. M. (2014). Tweets and trades: The information content of stock microblogsEuropean Financial Management20(5), 926-957.

Sprenger, T. O., Sandner, P. G., Tumasjan, A., & Welpe, I. M. (2014). News or noise? Using Twitter to identify and understand company-specific news flowJournal of Business Finance & Accounting41(7-8), 791-830.

Tumasjan, A., Sprenger, T. Sandner, P., & Welpe, I. (2011). Election forecasts with Twitter: How 140 characters reflect the political landscape. Social Science Computer Review29(4), 402–418.

Tumasjan, A., Sprenger, T. Sandner, P. & Welpe, I. (2010). Predicting elections with Twitter: What 140 characters reveal about political sentiment. Proceedings of the 4th International Conference of Weblogs and Social Media, Washington, D.C.

Project Team

Silja Spreng, M.Sc.

Dr. Barbara Stolz

Prof. Dr. Reiner Braun

Prof. Dr. Andranik Tumasjan