Skip to content

@ Conversations: A Rehearsal of “Beyond Microblogging”

7 July 2009

In a recent study at Indiana University, researchers uncovered how “@” signs function in Twitter as an indicator of conversational interaction. With the understanding that an @ sign is a form of addressivity, where a user indicates an intended addressee by typing that person’s name at the beginning of an utterance, researchers hypothesized that an increasing number of Twitter users employed them to signal their interaction with others.  Specifically, researchers set out to answer five questions to demonstrate the increasingly popular appropriation of Twitter as a conversational platform.

Research Questions:

1.    “What is the breakdown of the language of tweets across time periods, and to what extent is the @ sign used in tweets in different languages?”

Takeaway: It’s not surprising that English tweets were the most common of all the tweets captured on the public timeline, simply because the origin of the platform and the time bias of the tweets captured. What is interesting, however, across language groups the usage of @ signs was consistently about 30%.

2.    “How do instances of the @ sign function in English tweets?”

Takeaway: The most common function of the @ sign was ‘addressivity;’ nearly 91% of the tweets in the sample used the symbol to direct their tweets to another person. From this finding, it can be inferred that tweets that include an @ sign indicate the presence of a conversation on Twitter.  Following addressivity, the next most popular function categories were ‘reference’ and ‘location,’ accounting for 5.43% and 0.75% respectively. Taken together, the top three most common functions of the @ sign accounted for 97% of the usage.

Functional Categories:

•    Addressivity: directs a message to another person
•    Reference: Makes reference to another person, but does not direct a message to him or her
•    Emoticon: Used as part of an emoticon
•    Email: Used as part of an email address
•    Locational ‘at’: Signals where an entity is located
•    Non-locational ‘at’: Used to represent the preposition ‘at’
•    Other: Uses not fitting into any other category, including representations of swear words

3.    “What do people twitter about, and does it vary for tweets with and without @ signs?”

Takeaway: The most popular content of tweets, regardless of @ sign inclusion or exclusion, was reporting one’s own experience.  This finding is consistent with the stated purpose of Twitter to answer the question, “What are you doing?” What’s more interesting is that those tweets with an @ sign had a more diverse range of content themes. 33% of tweets with an @ sign  were about the addressee, 17% were self-expressive (compared to 51% of tweets w/o @ sign), and 16% included information for others.

Content Themes:

•    About addressee: solicits or comments on information relating to the addressee
•    Announce/advertise: announces information to the general readership of Twitter
•    Exhort: directs or encourages other(s) to do something
•    Information for others: posts information apparently intended for others; solicited or volunteered
•    Information for self: posts information apparently intended for sender’s own use
•    Metacommentary: comments on Twitter or twittering
•    Media use: reports or reflects on media use, especially music
•    Opinion: asserts a subjective or evaluative position
•    Other’s experience: solicits, reports on, or comments on info relating to the experience of a others
•    Self experience: reports or comments on sender’s own experience
•    Solicit information: requests information (other than about addressee)
•    Other: miscellaneous other themes (greetings, nonsense)

4.    “To what extent do English @ messages that are directed to others receive responses, either with or without @ signs?”

Takeaway: Of the public tweets in English that used an @ sign, 31.2% received a public response. This finding, albeit being a conservative estimate, shows the potential of Twitter to be an increasingly suitable conversational environment. Note: if the study was able to capture private tweets, as well as those replies that were posted an hour after the initial tweet, the percentage response may have been higher.

5.    “How long, and how coherent, are interactive exchanges, and to what extent do they make use of the @ sign?”

Takeaway: “On average in the sample, the conversations that occurred in Twitter appeared to be dyadic exchanges of three to five messages sent over a period of 15-20 minutes.”

Range Mean Median Mode
# Participants 2-10 2.5 2 2
Length of Exchange 00:25 – 54:22 17:41 26:33
# Exchanges 2 – 30 4.62 3 3
% Tweets Using @ per Exchange 20% – 100% 86% 100% 100%
Avg Time b/w Exchanges 00:25 – 34:05 06:43 04:24
General Methodology:

•    Four one-hour samples collected at four-hour intervals
•    Began collecting at 6 a.m. & ended at 6 p.m. EST on January 11, 2008
•    Used Twitter Scraper to capture tweets in three-second intervals & save to database
•    Note: Specific methodology varied from question to question

Time # Tweets % Times Tweets Missed
6 AM 7,754 5.59
10 AM 12,234 28.29
2 PM 8,471 16.29
6 PM 8,528* 8.79

*Questions 2-5 were answered based on the number of English tweets (1,472) within this sample

Overall, I found this study interesting because it offered ways in which we can further measure the interactivity of brands/people on twitter.  In addition to simply measuring RTs, we may now want to think about incorporating a content analysis of @ replies in our study of effectiveness and stickiness of Twitter users and their tweets.


Herring, Susan C and Courtenay Honeycutt (2009). “Beyond Microblogging: Conversation and Collaboration via Twitter.” Proceeding of the Forty-Second Hawaii International Conference on System Sciences (HICSS-42). Los Alamitos, CA: IEEE Press.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: