Full length articleWorking with tweets vs. working with chats: An experiment on collaborative problem solving☆
Introduction
Twitter is a free web platform enabling people to publish short messages, called “tweets”. When a new tweet is sent by a Twitter user, all her registered “followers” get notified. Since its debut in 2006, Twitter users have grown over half a billion, with 135,000 new users signing up daily and 190 million unique site visitors every day (Statistic Brain, 2014). Twitter has a very peculiar trait: each “tweet” has a maximum length of 140 characters. Therefore it can be considered as a “restricted length” communication medium.
Restricted length communication is not yet well known in organizations. Most firms and institutions mainly use Twitter as a new form of communication with the outside world. But Twitter is not yet largely adopted for collaboration and teamwork inside organizations. Is “working with tweets” actually possible? In what specific forms? Existing theoretical and empirical research on similar textual communication media, like chats (Instant Messengers: IMs, e.g. ICQ, AOL, MSN, Yahoo!, IRC, and Skype instant messaging), may represent a good departure point for an exploratory investigation. Extant research shows that IMs, while not a substitute for face to face interaction, have an important role in computer supported collaborative work. We do not know whether these findings could be extended to Twitter as well, and we aim to investigate whether working with tweets is fundamentally different than working with chats.
When attempting to apply the extensive literature on instant messengers to Twitter and restricted length communication media, important issues would arise.
First, there may be radical technical differences between tweets and textual chats. The most evident seems to be the restricted message length of Twitter. This limitation may produce important effects on conversational language and structure, as we already know from studies on individual communication by texting devices (SMSs). Besides message length, subtle differences in technical features like user interface, communication delay, user reach, and text persistence could importantly alter user perceptions, expectations, intents and actual usage patterns of restricted length communication media. We investigate the relevance of these technical differences as for a typical work activity, i.e. collaborative problem solving.
Second, in absence of specific theories on restricted length media, traditional concepts in CSCW like media richness and media synchronicity may be adapted to make sense of “working with tweets”. This exercise is delicate, and it would require taking into account technical, behavioural and even linguistic aspects, like expression form and discourse management strategies. We address this issue in the theoretical background session, reinterpreting consolidated studies on CSCW to understand what role Twitter and other restricted-length media may actually play in collaborative problem solving at work.
Third, despite the important, and growing, presence of Twitter in organizations, we still have to wait for years before seeing massive texting users (generation Y) at work. They are simply too young now. We recurred to laboratory sessions to observe young generations at work in collaborative problem solving exercises based on short messages. In the laboratory we can show how Twitter's 140 characters limit and technical features actually affect textual communication in teamwork, observing different teams at work, communicating by otherwise identical instant text message systems with and without the message length restriction and technical features typical of Twitter.
The experiment proposed here is focused on textual-only communication for group problem solving, comparing Twitter, the 140-character microblog, with the Skype feature of instant messaging. Skype is one of the first and most successful synchronous communication and collaboration platforms.1 Both application are main players and widely diffused; both are available for free for all the main technological platforms and operating systems.
The comparison of Twitter and Skype instant messaging (Skype IM) was accomplished in a series of laboratory experiments in which a problem solving activity was carried out by groups of subjects using the two different systems.
Given the 140 characters message length restriction, we expected to find significant differences in communication patterns between Twitter and Skype IM sessions. Our results confirmed these expectations: the average Twitter users exchanged a significantly lower text volume, measured both in characters and in message count.
These outcomes are in resonance with recent research findings, suggesting that lower degree of media richness and synchronicity, together with the higher rehearsability and potential of public access, in presence of the 140-character limit, could push users to a lower degree of “chatness” reducing the volume of exchanged messages.
More surprisingly, the actual message length resulted higher in Twitter than in Skype IM sessions. This result, only apparently in contrast with the 140 characters message length restriction in Twitter, is commented in the Discussion Section.
On the other hand, in our experiment, Twitter problem solving performance, was not negatively affected by the lower text volume exchanged: On average, Twitter groups performed just as well as Skype IM groups.
This result, while limited and partial, supports the view of restricted-length lean media as effective means in low intensity communication, suggested by recent studies reviewed below.
The structure of the paper is the following: next section presents a brief historical evolution of Twitter diffusion and use. Section 3 proposes a theoretical discussion of restricted length media peculiarities; exploring and interpreting extant literature from the technical, behavioural, and linguistic point of view, and setting grounded expectations for an empirical exploratory analysis. Section 4 is dedicated to the experimental research methodology. The experimental focus and the setting, the data sample and measurements are discussed. Section 5 reports the results of the experiments. Section 6 provides a discussion of our findings. Finally, the last section reports some concluding remarks and limitations of the study and suggest further experimental analyses for future investigations.
Section snippets
Twitter usage: previous studies
Twitter was created in October 2006 as a free service that allows users to communicate via text-based messages of up to 140 characters known as ‘tweets.’ Twitter is a microblogging service where users post status messages and short communications (a.k.a., tweets) to a network of associates (a.k.a., followers) from a variety of devices. Tweets are text-based posts of up to 140 characters in length. The default setting for tweets is public, which permits people to follow others and read each
Understanding Twitter usage: theoretical background
Our research may be positioned within the study of Social Network Sites and tools (SNS), an emerging area opening new and exciting fields of exploration including the study of usage patterns, the availability and analysis of the so called “big data” produced by social networks and the design of new organizational forms and new business models enacted by social networks (Beer, 2008, Boyd and Ellison, 2007). By analysing the usage of Twitter in workgroup activities, in comparison with a typical
Methodology
The design is based on randomized experiment (cfr. e.g. Shadish, Cook, & Campbell, 2002). It is particularly appropriate to this investigation, not only for the impossibility to currently observe “generation Y” employees in field studies (younger generations are simply too young to work in firms), but also to explicitly take into account the influence of the dimensions under investigation (alternative media choice by different subjects, tasks, organizations), controlling any other external
Experimental setting
In the series of binary experiments reported in this paper, users have been randomly assigned to two groups of the same size working in the same conditions to complete a certain problem solving task. Group A (“treatment”) is requested to use the textual communication tool under observation, limited to 140 char (Twitter). Group B (“control”) is required to use a similar, but unlimited-length textual communication tool (Skype instant messaging).
According to this experimental design (cfr. e.g.
Results
Our output results take into account group problem solving performance, message volume and frequency, actual average message length, and users' perceptions in the adoption choice.
Discussion
From the experimental comparison of restricted length lean communication (Twitter) with unrestricted length instant messaging (Skype IM) in collaborative problem solving the following aspects emerged: 1) For both systems the average message length was much shorter than Twitter's limit of 140 characters, and it was surprisingly higher in Twitter than in Skype IM (about 28 vs. 20 characters per message); 2) Skype dominated Twitter IM as for the total volume of text exchanged, message count, and
Conclusions
Is really “working with tweets” a viable and useful prospect for the next “Generation Y” organizations? Our experimental approach investigates how Twitter and restricted lean media may affect organizations in the near future.
Our first result is that the 140 characters limit actually affects teamwork communication patterns: Twitter teams interacted significantly less than Skype IM users, exchanging less messages and less frequently. According to this experimental result, the 140 characters limit
Acknowledgements
Many thanks to Roberta Mion for her research assistance at the Ca' Foscari University of Venice, and to the students of the “dream team” (Fabio Carrocci, Valerio Ionta, Mario Panaccione, Enzo Prata, Luca Savo Santone), who gave concrete support in the organization of the experimental sessions at the University of Cassino. We kindly acknowledge Maria Rees, assistant professor at the Ca’ Foscari University of Venice, for her English-language support. We are grateful to the editorial, review and
References (67)
- et al.
Computer-mediated communication and group decision making: a meta-analysis
Organizational Behavior and Human Decision Processes
(2002) - et al.
A critical assessment of potential measurement biases in the technology acceptance model: three experiments
International Journal of Human-Computer Studies
(1996) - et al.
Is that a bot running the social media feed? Testing the differences in perceptions of communication quality for a human agent and a bot agent on Twitter
Computers in Human Behavior
(2014) - et al.
The poverty of media richness theory: explaining people's choice of electronic mail vs. voice mail
International Journal of Human-Computer Studies
(1997) - et al.
Texting versus talking: an exploration in telecommunication language
Telematics and Informatics
(2013) - et al.
Comparing the task effectiveness of instant messaging and electronic mail for geographically dispersed teams in Taiwan
Computer Standards & Interfaces
(2007) - et al.
Users of the world, unite! the challenges and opportunities of social media
Business Horizons
(2010) Can communication medium limitations foster better group outcomes? An action research study
Information & Management
(1998)- et al.
If you are quick enough, I will think about it: Information speed and trust in public health organizations
Computers in Human Behavior
(2014) - et al.
What kind of #conversation is Twitter? Mining #psycholinguistic cues for emergency coordination
Computers in Human Behavior
(2013)
Using mobile communication technology in high school education: motivation, pressure, and learning performance
Computers & Education
Impact of communication medium on task performance and satisfaction: an examination of media-richness theory
Information & Management
Information-seeking behavior in generation Y students: motivation, critical thinking, and learning theory
The Journal of Academic Librarianship
Perceived usefulness, ease of use, and usage of information technology: a replication
MIS Quarterly
Communication characteristics of instant messaging: effects and predictions of interpersonal relationships
Instant messaging and the future of language
Commununications of the ACM
Always on: Language in an online and mobile world
Social network(ing) sites: revisiting the story so far: a response to Danah Boyd & Nicole Ellison
Journal of Computer-Mediated Communication
Social network sites: definition, history, and scholarship
Journal of Computer-Mediated Communication
Team size and technology fit: participation, awareness, and rapport in distributed teams
IEEE Transactions on Professional Communication
Business and web 2.0: An interactive feature
Discourse management in three modalities
Language@Internet
A coordination theory approach to organizational process design
Organization Science
How large U.S. Companies can use Twitter and other social media to gain business value
MIS Quarterly Executive
Information richness: a new approach to managerial behavior and organizational design
Research in Organizational Behavior
Organizational information requirements, media richness and structural design
Management Science
The multilingual Internet: Language, culture, and communication online
Perceived usefulness, perceived ease of use, and user acceptance of information technology
MIS Quarterly
User acceptance of computer technology: a comparison of two theoretical models
Management Science
Media, tasks, and communication processes: a theory of media synchronicity
MIS Quarterly
Testing media richness theory in the new media: the effects of cues, feedback, and task equivocality
Information Systems Research
A social influence model of technology use
Trust and TAM in online shopping: an integrated model
MIS Quarterly
Cited by (10)
Assessing verbal interaction of adult learners in computer-supported collaborative problem solving
2023, British Journal of Educational TechnologyLiterally Looking at Links – Mapping Communication and Content on Twitter
2022, Communications in Computer and Information ScienceCollaborative problem-solving: How to implement and measure it in science teaching and learning
2021, Journal of Physics: Conference SeriesDrinking from the firehose: The structural and cognitive dimensions of sharing information on twitter
2020, American Journal of EducationA metric to assess brokerage positions within social networking sites
2020, Teachers College Record
- ☆
All the authors worked in close collaboration in researching and writing this study, and their names are in alphabetical order. Authorship should be attributed as follows: Daniela Isari sections 2,4,5,6; Andrea Pontiggia sections 1,8: Francesco Virili sections 3,7.