Online Political Communication Strategies: MEPs, E-Representation, and Self-Representation

ABSTRACT

Research into the communication strategies of legislators has a long history. The European Parliament offers an opportunity to add to understanding of how legislators prioritize styles of communication, with a comparative perspective across 27 nations. Through content analysis of online communication, we investigate how the Internet is used by members of the European Parliament. Our analysis assesses three communication strategies: homestyle, impression management, and participatory. We find that a homestyle strategy predominates, followed by impression management. Participatory communication is emergent, but may earn legislators political capital, as it appears that proactive communicators who offer participatory opportunities are more likely to build an online following.

Keywords:

• Homestyle
• impression management
• interactivity
• legislators
• representation
• social media
• strategic communication

Notes

1. MEP profiles are provided on http://www.europarl.europa.eu/members/expert/groupAndCountry.do?language=EN [entry 02.11.2011]. Only 440 out of 736 MEPs had official Web sites. Additionally we searched Google.com for the remaining 296 MEPs by entering name and surname (we found that probably 166 of them had Web sites; however, these sites were not linked to their EP profiles). Due to methodological issues (e.g., we were unsure that the Web sites we found on google.com are official MEP Web sites, as in some case there was more than one Web site listed, some had not been updated since 2009, or some were not listed among the first 50 entries due to the MEP having a popular surname). To avoid all the problems caused, we decided to exclude from the analysis all those MEPs who did not officially list their Web site addresses.

2. November was chosen as a “normal” month, with MEPs into a daily routine after their summer holidays but not close to the Christmas break.

3. The data archives were downloaded to a local computer at Sciences-Po, Paris. This was performed by TelePort Ultra, which was provided by Tennyson Maxwell Information Systems, Inc.

4. We decided to keep this strategy regardless of the low Cronbach score, as George and Mallery (2009, p. 231) recognize this score as “poor.” However it is consistent with our theoretical assumption of the HIP, as well as with the existing literature on legislators use of the Internet.

5. In this paragraph we use outcomes from the logistic regression where the dependent variable was having = 1 or not having = 0 an official link from the EP profile to the MEP's Web site. All numbers in the brackets are odds ratios with statistical significance: *p < .10, **p < .05, ***p < .01.

6. Poisson regression was chosen as the best statistical method for estimating count data variables (Wooldridge, 2001, p. 645).

7. The multilevel regressions were run for each strategy for model of different level variables [micro (individual), mezzo (EP parliament), and macro (country), not shown due to the lack of space, available from the authors]. However no interesting differences were found (with the exception for gender difference influence on Impression Management Strategy, as women were less likely to use it, however the effect was lost when introducing EP and country characteristics). As expected after introducing new set of variables, R² rose, achieving the final level indicated in Table 2.

8. Multilevel regressions models were run; however as there are no significant differences, only the final model is shown.

9. Out of 440 MEP under investigation.

10. Originally, the in-depth interviews and quantitative surveys with MEP and office members were a part of this research project; unfortunately the response rate of the MEP offices was at the 3% level.