Predicting opinion leadership on twitter

Predicting Opinion Leaders in Twitter Activism Networks: The Case of the Wisconsin Recall Election

Weiai Wayne Xu (Univ. at Buffalo)

Yoonmo Sang (Univ. of Texas-Austin)

Stacy Blasiola (Univ. of Illinois at Chicago)

Dr. Han Woo Park (YeungNam University, S. Korea)

Presentation for #SMSociety2014 (September 27-28, Toronto)1

1. Networked opinion leadership

2

• Wisconsin recall election

• #wirecall

• User-to-user follows relationship

• On Twitter, opinion leadership

means getting your message

retweeted.

2. Our goal

3

• _____?

• _____?

• _____?

• What user characteristics and

behaviors predict opinion

leadership on Twitter?

3. Classic opinion leadership model (Rogers, 2003)

4

• Social connectivity

• Involvement

• Knowledge

• Status

• Etc.

Rogers, E. M. (2003). Diffusion of innovations (5th ed. ed.). New York: Free Press.

Are these attributes still relevant in digital age?

4. Linking the opinion leadership model to Twitter

5

• Social connectivity

• Twitter forms information flow networks through follows, retweet and mention.

• A higher betweenness centrality is indicative of a higher level of connectivity

4. Linking the opinion leadership model to Twitter

6

• Involvement, knowledge, status?

4. Linking the opinion leadership model to Twitter

7

• Involvement, knowledge, status?

action

community

information

Explicitly ask other users to engage in certain acts

Providing original feedback, interactive

Non-directed, one-to-many, simply passing along

others’ messages

Engagingtweets

5. Key hypotheses

8

• Hypothesis 1: Users’ centrality in Twitter networks is related to

influence on the diffusion of political information such that the higher

the centrality, the more likely users’ messages are retweeted by other

users.

• Hypothesis 2: The more politically involved the users are, based on the

level of self-disclosure of personal political information, the more likely

users’ messages are retweeted by other users.

• Hypothesis 3: The more involved the users are in a given political issue,

based on their geographic proximity to the political event, the more

likely their messages are retweeted by other users.

• Hypothesis 4: The more involved the users are in a political issue,

based on their contribution of engaging tweets, the more likely their

messages are retweeted by other users.

In short, we hypothesize that more connected and involved users are more successful in influencing information flow within Twitter networks.

6. Data collection

9

• Most recent 1500 tweets every two hours, from 5-29-2012 to 6-

5-2012.

• 1000 users randomly sampled from 8957 Twitter users that

tweeted #wirecall during the timeframe

• The sampled users sent 3546 tweets containing the hashtag

#wirecall

7. Results

10

The results provided general support for the hypotheses:

• The model explained 26% of the variance, F(6,593) = 8.22, p< .001.

• Betweenness centrality was positively related to the number of RTs

(β = .26).

• local users were more likely to be retweeted (β = .20).

• issue involvement based on engaging tweets (β = .21) positively

predicted the number of RTs.

• political involvement DOES NOT predict RT.

8. Takeaway

11

Opinion leadership in social media is contingent upon both

network and context factors.

• Characteristics associated with traditional opinion

leaderships are still relevant in Twitter communication.

• Integrating network analysis and content analysis

9. Future directions

12

• Combining behavior data and perception data (content

analysis + network analysis + survey)

• Connectivity in various types of networks (issue network

vs. general Twitter network)

• Non-issue specific

• Longitudinal analysis

Contact

13http://abs.sagepub.com/content/58/10/1278

• Weiai Wayne Xu: [email protected] http://curiositybits.com/

• Yoonmo Sang: [email protected] http://rtf.utexas.edu/graduate/phd-

year-4

• Stacy Blasiola: [email protected] http://blasiola.wordpress.com/

• Dr. Han Woo Park: http://www.hanpark.net/