The Credibility of Digital Identity Information on the Social Web: A User Study
Download as PPTX, PDF3 likes542 views
The document summarizes a study that assessed the credibility of digital identity information on social web platforms by comparing users' real-world and digital social networks. It found that on average, users' digital networks covered 77% of their real-world networks, consisting mostly of strong ties like friends. While digital networks contained some weak ties, behavior analysis showed users mainly interacted frequently with a small number of close contacts from their real-world networks. Therefore, digital identity information on social media can be considered reasonably credible representations of users' real-world social identities and relationships.
The Credibility of Digital Identity Information on the Social Web: A User Study
- 1. The Credibility of Digital Identity Information on the Social Web: A User StudyMatthew RoweOrganisations, Information and Knowledge GroupUniversity of Sheffield
- 2. OutlineProblem: Increased reliance on digital identity informationMotivation: Need for assessing the credibility of digital identity informationDigital IdentityThe tiers of identity informationComparing Digital and Real-World IdentitiesUser study of Social Web usersFindings from the StudyConclusions
- 3. ProblemThe World Wide Web has evolved from a Web of anonymity into an accountable WebUsers are able to interact online with Web platformsPublish their thoughts and musings on blogging platforms such as LiveJournalUpload videos onto YoutubeShare photos on FlickrChat with friends on MySpaceAll of the above allow users to construct bespoke online personasSo called “Digital Identities”Shaping how they wish to be perceived by others in an online environment
- 4. MotivationDigital identity information from Social Web platforms is used by a range of 3rd party applications and services:Dopplr uses social network information from user profiles to share travel arrangements and planned tripsGoogle’s Social Search uses identity information to affect search listingsIdentity management services have begun using digital identity information Social Web platforms to support automated identity disambiguation techniquesEmployers use Social Web platforms to perform lateral surveillance of potential and current employeesThe increased reliance on digital identity information from Social Web platforms requires an assessment to be made of the credibility of this informationTo what extent is real identity information mirrored in a digital space?
- 5. Digital IdentityThe Oxford dictionary defines identity as:“the fact of being who or what a person is” and…“the characteristics of defining this”A person’s identity is comprised of a set of attributes which makes them uniqueDigital Identity reflects this, however in an online space the these attributes can be customised and tailored by the userThe functionalities and feature sets provided by Social Web platforms facilitate this alterationUsers can construct profiles containing their biographical information (name, address, email) along with their social network information
- 6. Digital IdentityDigital Identity can be divided into 3 tiers:My Identity: persistent identity information such as name, date of birth and genealogical relationsShared Identity: information which is susceptible to change such as social network informationAbstracted Identity: identity information derived from groupings and demographics(Ploderer et al, 2008) states that self-promotion motivates web users to create a profile(Lampe et al, 2006) found Social Web users to search for and interact with those people which they knew offlineThe social network which a web user maintains is a powerful tool in establishing a unique identification of the personThis is used by identity disambiguation techniques to confirm person references on the Web
- 7. Comparing Digital and Real-World IdentitiesSocial network information from the shared identity tier provides a means of quantifying the similarity between digital and real-world identities:To what extent is real identity information mirrored in a digital space?To explore this question a user study was conducted using the Social Web platform FacebookGiven that this is the most popular Social Web platform in the UK22 million users in the UKhttp://www.clickymedia.co.uk/2009/10/uk-facebook-user-statistics-october-2009/50 participants were gathered from the University of Sheffield25 male and 25 femaleAge range of 18-45The experiment was conducted in November 2008
- 8. Comparing Digital and Real-World IdentitiesThe user study consisted of 3 stages:Each participant listed their real-world social networkUsed a web page form with 20 rows, one for each person of the network, containing their name and the relationship typeMore rows could be added if neededContains strong-tied relationships (Donath & Boyd, 2004)Each participant’s digital social network was extracted from FacebookUsing an application designed for this studyAlso analysed the behaviour of the the participant: who they appear in photos with and who they share messages withEach participant compared his/her real-world and digital social networkSelected participants which appeared in both networks
- 9. Comparing Digital and Real-World IdentitiesTo measure the extent to which digital identity information mirrors real-world identity information the following evaluation measures were used:RelevanceAdapted from the information retrieval metric: precisionMeasures the ratio of strong-tied to weak-tied relationships in the digital social networkCoverageAdapted from the information retrieval metric: recallThe proportion of the real-world social network which appears in the digital social network
- 10. Comparing Digital and Real-World IdentitiesRelevanceAverage relevance measure of 0.23, indicating that 23% of a digital social network contains strong-tied relationshipsCoverageRanges between 0.5 and 1, with an average coverage of 0.77This indicates that, on average, 77% of a participant’s real-world social network appears onlineDifferent from findings by (Subrahmanyam et al, 2008) which found only 49% for coverage (they define it as overlap)Possibly due to differing dates when the studies were conducted and the differing countries
- 11. Comparing Digital and Real-World Identities
- 12. Comparing Digital and Real-World IdentitiesThe covered portion of the real-world social network in the digital network was also analysed, and showed the following:Majority of relationships were with friends (62%)24% were with family14% were with coworkersFor the different relationship types in the real-world social network89% of friends were replicated online73% of family were replicated online68% of coworkers were replicated onlineThe largest age group in the social network that was covered online was <21Indicative of the sample used by the study
- 13. Comparing Digital and Real-World IdentitiesBehaviour of each participant was analysedAssessing who they appeared in images with and interacted with on the platform Results demonstrate a tendency to interact with only a few people frequentlyAnd a large number of people rarelyThe graph forms a power law curveDenoted as the Social Longtail (>15)The longtail contains 92% of strong-tied relationships which appeared in the real-world network listed by participantsSuggests that offline interactions and communications are continued onlineDemonstrated by the number of messages sent and photos in which both people appear
- 14. ConclusionsCredibility of digital identity information is important to many application which rely on such informationFindings from the study quantify the significant extent to which web users replicate their real-world identities in an online spaceCharacterised by the large coverage of real-world social networks on an example Social Web platformAn insight is provided into the credibility of digital identity information found on the Social WebBehavioural trends demonstrate the continuation of interaction offline in a digital environmentThe Social Longtail is mainly comprised of strong-tied relationships found offline, driven by frequent interactionsThe results also provide findings in contrast to (Subrahmanyam et al, 2008)Due to the different domains in which the studies were conducted
- 15. Twitter: @mattroweshowWeb: http://www.dcs.shef.ac.uk/~mroweEmail: m.rowe@dcs.shef.ac.ukQuestions?(Donath & Boyd, 2004) - J. Donath and D. Boyd. Public displays of connection. BT Technology Journal, 22(4):71–82, October 2004.(Lampe et al, 2006) - C. Lampe, N. Ellison, and C. Steinfield. A face(book) in the crowd: social searching vs. social browsing. In CSCW ’06: Proceedings of the 2006 20th anniversary conference on Computer supported cooperative work, pages 167–170, New York, NY, USA, 2006. ACM.(Ploderer et al, 2008) - B. Ploderer, S. Howard, and P. Thomas. Being online, living offline: the influence of social ties over the appropriation of social network sites. In CSCW ’08: Proceedings of the ACM 2008 conference on Computer supported cooperative work, pages 333–342, New York, NY, USA, 2008. ACM.(Subrahmanyam et al, 2008) - K. Subrahmanyam, S. Reich, N. Waechter, and G. Espinoza. Online and offline social networks: Use of social networking sites by emerging adults. Journal of Applied Developmental Psychology, 29(6):420–433, November 2008.