Scientists online: A framework for the analysis of Internet profiles
First Monday

Scientists online: A framework for the analysis of Internet profiles by Helena Bukvova

Many scientists use the Internet to present themselves and their work. The content they create could be used to improve the awareness and communication within the scientific community. This requires a sound understanding of the contents on scientists’ profiles, especially with regard to their structure. Existing literature offers mostly basic categorisation, focusing only on single platforms. This article presents a study of scientists’ profiles on institutional and private Web pages, social networking services, blogs, and microblogs. The aim of the study was to describe structures within the profile contents. For this purpose, 79 profiles belonging to 15 German scientists were identified and analysed using the constructivist grounded theory method. The result was a framework, suitable for structuring and further analysis of scientists’ profiles. The framework describes three levels for the study of profiles: profile networks, profile instances, and content units. The content on the profiles can be classified with regard to its type, verbosity, and placement. The developed framework serves as a basic structure for further research into scientists’ online self–presentation.






This article is a contribution to foster a better understanding of scientists’ self–presentation on the Internet. The Internet offers a number of platforms, where scientists can present themselves and their work (Hess, 2002). They can use simple HTML pages, profiles on social–networking–system (SNS) platforms, blogs, microblogs, wikis, for example. Many platforms which have previously focused on information management (e.g., bookmarking platforms or citation–management platforms) are now adding profiling and networking features. There are also platforms focusing explicitly on scientists. The content of such profiles is accessible to a wide audience. Scientists increasingly need to collaborate in international and interdisciplinary projects (Sonnenwald, 2007). They thus have to identify potential partners outside local or disciplinary networks. The online self–presentation of scientists can serve to improve the awareness of and communication within the scientific community (compare Genoni, et al., 2005; Farooq, et al., 2007). This potential could be further developed using information and communication technologies for example to identify suitable partners or to aggregate content from different sources (Carroll, et al., 2009).

The development of appropriate tools requires a sound understanding of the content on scientists’ online profiles, particular with regard to its structure. The existing research has explored mostly the online presence of general public (e.g., Evans, et al., 2008; Stern, 2004; Strufe, 2010; Grasmuck, et al., 2009). Studies of content created by scientists focus on single platforms, e.g. Web pages (Hess, 2002) or blogs (Ferguson, et al., 2010). Scientists can, however, have several profiles on different platforms. Thus a more holistic approach is needed. Furthermore, while some findings offer general categorisation and typologies (e.g., Herring, et al., 2005; Ewins, 2005), these are not specific enough to serve detailed content analyses.

In this article, I present a general model to structure content on scientists’ online profiles. The model was developed as a part of a larger research project to serve as a foundation for the identification of patterns in scientists’ online profiles. The design of the model was data–driven; 79 profiles belonging to 15 German scientists were analysed using the constructivist grounded theory method. As a result, I have developed a framework structuring scientist’s online presence with regard to three analytical levels: profile network, profile instance, and content unit; and three content dimensions: type, verbosity, and placement. This structure appears consistent with the existing research on online self–presentation. The framework is suitable for structural analysis of scientists’ online profiles.




The study methods were based on the principles of the constructivist grounded theory method (Mills, et al., 2006). The grounded theory method is “a systematic qualitative approach to data collection and analysis, that is concerned with generating theory” (Holloway and Todres, 2006). Theories are thus not deductively generated a priori and tested in an empirical setting, but constructed based on the data (Glaser and Strauss, 1967; Charmaz, 2000). Key features of grounded theory are a systematic, but highly iterative procedure of simultaneous data collection and analysis, based on constant comparison between already coded and new data samples (Charmaz, 2006). The data are analysed as they are gathered and further data collection is directed by the preliminary results (theoretical sampling, Morse, 2007).

I was interested in the profiling content of scientists, whom I defined as individuals professionally engaged in research at an academic institution. I have viewed scientists’ profiles on institutional Web pages, personal Web pages, SNS, blogs, and microblogs. Of interest were profiles connected to the individuals’ role as scientists. Private pages and content (e.g., family Web pages, hobby–oriented blogs) were not taken into account. The sampled data were sorted into cases. Each case described all identified and relevant profile pages belonging to an individual scientist. The cases were categorised according to five characteristics of the profile owner: (1) position; (2) sex; (3) research area; (4) country of the currently affiliated institution; and, (5) activity on particular platforms.

The study was carried out in three steps. Each step started with sampling of five scientists. The samples were chosen purposively to provide high heterogeneity and thus different patterns of profile information. However, the structure of the samples changed according to the needs of the study. For each scientist, I have viewed activity on the following platforms: personal profile on institutional pages (static HTML), SNS profiles (including SNS–like profiles, e.g., on citation–management platforms), blog, microblog, and any further static HTML profiles. A foundation for locating the different online profiles was the linking of the case individuals from their known blogs and microblogs. Additionally, active search was used to locate the institutional profile as well as further SNS profiles on Xing [1], LinkedIn [2], ResearchGate [3], and Academia [4]. Profiles from other platforms (e.g. Zotero [5], CiteULike [6], SlideShare [7]) were taken into account if they were linked to, but no active search was carried out on these platforms. A snapshot was taken of each located profile page either in PDF or in JPEG format for further analysis. In case of blogs, besides collecting all static pages, snapshots of the last 15 posts were taken. For microblogs, the last 50 posts were collected besides the profile page. The data collected from the different profiles were analysed and coded using the qualitative data analysis software AtlasTI [8]. First, the data were coded with terms or phrases used by the case individuals (in vivo coding, Charmaz, 2003; 2006). These first codes were then aggregated into higher–level codes and categories. The analysis process was highly iterative, including constant comparison with and recoding of already analysed data.

The first sample was heterogeneous with regard to (1) position; (2) sex; and, (3) research area of the scientists. The sample was homogeneous regarding (4) the location of the affiliated institution, all cases coming from Germany, and (5) platform activity, all cases being scientists who have at least a blog and a microblog. The activity as an author of a scientific blog and an owner of a microblog was assumed to be an indicator of a high level of online activity. The data were then iteratively coded and categorised. The result of the analysis was a first categorisation of the profiling information provided by the scientist. The second sample was heterogeneous with regard to (1) position; (2) sex; and, (3) research area. It was also homogeneous regarding (4) the location of the affiliated institution, all cases coming from Germany. The homogeneity regarding (5) platform activity was decreased by including researchers with blogs but not necessarily with a microblog. The reason for this change was to view cases with sufficient level of online engagement, but with higher diversity of platform activity than the first sample. The already identified categories were further specified using the second sample. The third, final sample was selected in the same way as the second sample. The analysis of the third sample data served to solidify the developed structure.


Table 1: Aggregated sample structure.
CharacteristicsRepresentation in the sample
(1) Position8 assistants, 7 professors
(2) Sex8 male, 7 female
(3) Research area6 from natural sciences,
5 from social sciences,
4 from linguistics and cultural sciences
(4) Institution locationGermany
(5) Platform activityAll with a blog, some also microblog


Altogether, I have analysed profiles belonging to 15 scientists. Table 1 shows the structure of the aggregated sample. Each scientist had several profiles on different platforms. In total, 79 profiles were examined and coded. The profiles came from different platform types, including static HTML (private or institutional), SNS, blogs, microblogs, and other platforms that support profiling (e.g., citation–management platforms). The distribution of the platform types is shown in Figure 1.


Figure 1: Distribution of platform types
Figure 1: Distribution of platform types.


When collecting data for analysis, ethical issues were also considered. Data gathered through indirect observation, particularly in a virtual environment, can be collected without the subjects’ awareness. Although all data on the platforms of interest were publicly accessible and thus technically in a public domain (Herring, 1996; Whiteman, 2007), they were still the intellectual property of the authors. As such, they have been created with a certain purpose and the authors may not be comfortable with the use of their data for research purposes (Bakardjieva and Feenberg, 2000). Furthermore, collecting data from the platforms involved recoding of personal data. The anonymisation (Frankfort–Nachmias and Nachmias, 2008) of the data and at the same time the maintenance of the depth required for the analysis was practically impossible. Therefore, all scientists, whose data were recorded, were contacted per e–mail, informed about the study, and asked for permission to use their content (informed consent, King, 1996; Bakardjieva and Feenberg, 2000; Mitchell, 1993). Scientist who have refused their consent or who did not replied were excluded from the study. This procedure might have introduced some bias to the sample, as four subjects from similar backgrounds refused consent and had to be replaced by alternative cases to obtain the final sample.

Using the described procedure, I have structured the data first with simple codes, later with more abstract categories. The codes and the categories were constantly reflected. After the third step, I have considered the developed structure sufficiently stable to serve as an analytical framework.




The result of the study is an analytical framework, a model for the structuring of content found on scientists’ profiles. The model was designed to be independent of platform type. The content on scientists’ profiles can be studied on three levels: content unit, profile instance, and profile network. Depending on the level, it can structured further according to three dimension: type, verbosity, and placement. The framework (see Figure 2) is described in detail in the following sections.


Figure 2: Resulting analytical framework
Figure 2: Resulting analytical framework.


Analytical levels

The sampled scientists had a high level of engagement and their online presence was complex, spanning across several platforms. Even on a single platform, the scientists often provided a large amount of content. It was thus necessary to analyse a scientists’ online presence as a multilevel structure. I have identified three levels of analysis:

Profile instance is defined as the content provided as a part of a single profile on a particular platform. It is the central level. Profile instances can contain several pages.

Profile network is a network of profile instances belonging to one scientist. The profile instances can be connected with hyperlinks. A profile network can thus be depicted as a directed graph.

Content units describe packages of content defined for analytical purposes. Complex profile instances are difficult to analyse. It is therefore of advantage to divide the content on a profile instance into units with similar characteristics. An information unit belongs to one profile instance, but it can span several pages within the profile. Its exact definition is directed by the analytical needs.

Combined with the analytical dimensions, the three levels support a structured, complex exploration of scientists’ profiles (see Figure 2). The content units can be described with regard to their content type and verbosity. The results of such analysis can be aggregated on the profile–instance level and combined with information about the platform type of the profile instance. Furthermore, the placement of the profile instance within the profile network and in relation to other profile instances can be considered. The placement of all profile units within a profile network can be aggregated to study patterns in the network.

Type of content

This dimension described what types of content researchers provide about themselves on the different platforms. I have identified four content types: (1) identification; (2) activities; (3) achievements; and, (4) expertise. The identification content connects the profile to its owner. Content about activities, achievements, and expertise describes the scientists’ professional engagement regarding academic work, teaching, research, and personal qualification. These three types are connected: past activities can be referenced as achievements, achievements lead to expertise, and based on their expertise the scientists engage in further activities. As a part of professional profile, the activities, achievements, and expertise again influence the scientists’ identification (see Figure 3). I have defined the content types as follows.


Figure 3: Identified content categories
Figure 3: Identified content categories.


Identification. Content of this type provided information necessary for linking the profiles in the virtual world to the real–world scientists. In the first place, this included the description of the individual behind the profile, such as name, photo, or interests. Furthermore, on professional profiles, the identification content also described the profile owner as a scientist, including affiliation information. Finally, direct contact to the profile author also provides a link to the real–world individual, so that any contact data, online or off–line, could be seen as identification content.

Activities. The scientists in the sample were engaged in a number of professional activities and could provide information about these activities as a part of the profile. These could be related to their (1) research work (e.g., information about current research projects, research interest, working papers); (2) their official or unofficial work within the scientific community (e.g., information about conference organisation and attendance, reviewer activity, editor activity, professorships); (3) their work in teaching (e.g., current lecturing activities); or, (4) personal development (e.g., current further qualification). Activity–related content gave information about things that currently occupied the scientists. Once the activities were finished, they became a part of the scientists’ achievements.

Achievements. Throughout their career, the scientists accumulated information about activities in which they have (successfully) engaged. Completed activities became achievements and could be used by the scientists as references. They could be provided as a part of their profiles. Very typical was the presentation as a curriculum vitae. As achievements were derived from activities, they could also be categorised into (1) research achievements (e.g., publications, finished projects); (2) academic achievements within the scientific community (e.g., academic career, past work as reviewer or editor, past conference organisation, past academic positions); (3) teaching achievements (e.g., taught lectures); and, (4) personal development (e.g., attained degrees, attended courses, certificates). Through their achievements, the scientists accumulated special expertise.

Expertise. The scientists possessed expertise which was derived from their activities and achievements. It could include knowledge or skills related to their (1) research work (e.g., methodological skills, specialised IT skills, knowledge in particular areas); (2) academic work in the scientific community (e.g., reviewing expertise, organisational skills); (3) teaching (e.g., pedagogical skills, teaching knowledge); or, (4) personal development (e.g., language skills). Information about expertise could be explicitly stated or implied through active demonstration of the possession of particular knowledge or skills. The possession of expertise could lead to further engagement in specific activities.

A profile instance typically contained different content types. For a deeper analysis the instance can be structured into content units. The definition of the units depends on the aims of the analysis. For example, content could be divided based on its type, allowing for aggregation of content of the same type into content units. Or it would be possible to create content units according to structural criteria (e.g., the content’s position on a Web page, to study what content is provided first). This would then result in multi–type units.


Besides noting the types of the contents on the scientists’ profiles, it was further necessary to assess the verbosity, i.e., the amount of content provided by the scientists concerning a particular topic. Verbosity could be measured by simple quantitative means, e.g. number of words, and thus inform about the length of the profile. However, interpreted and evaluated as a qualitative measure, it helps to understand how the scientists develop the content. For this purpose, verbosity can be assessed with regard to three categories: (1) the amount of factual information provided; (2) the level of personalisation; and, (3) the level of interaction.

Factual information. This first category of verbosity describes the amount of factual information that the scientists have written about the particular topic. It ranged from basic facts, through further information, to elaborate narrative descriptions of the context. The amount of factual information provided could be described using six levels:

  1. None. No information provided about a particular topic.
  2. Implicit. The information about a particular topic were not provided directly, but they could be deduced from the context, e.g., from information provided about other topics.
  3. Noted. Some information were provided, however, they are not sufficient to fully establish all the facts.
  4. Stated. Sufficient basic descriptive information were provided, often in a structured way.
  5. Descriptive. Facts and necessary context were provided, often in a narrative form.
  6. Detailed. Elaborate, usually long, narrative description was provided, discussing facts as well as the context.

Personalisation. This second category describes the amount of information given by the scientists about their personal position in relationship to the particular topic. Besides providing facts about a topic, some scientists chose to reveal their opinions, thoughts, or ideas. Again, the personalisation could range for basic statements to detailed discussions. The level of personalisation could be described using four levels:

  1. None. No personalisation of a particular topic.
  2. Personal notes. The profile instance contained few remarks about the individual’s personal relationship to the topic (e.g., experiences, opinions, thoughts, humour, likes, dislikes). The focus remained on the facts.
  3. Personalised. The focus was mostly on the facts, but the relationship of the author to the topic formed an important part of the content.
  4. Highly personalised. The focus was on the individual’s relationship to the topic. Facts were provided to give the audience background necessary to appreciate the individual’s argumentation.

Interaction. The third category describes the scientists’ willingness to interact with their audience regarding a particular topic. This could be demonstrated in the text itself, e.g., through direct addresses of the audience, or in further measures taken by the researchers, e.g., participation in discussion about the topic. The level of interaction could be described using four levels:

  1. None. No interaction about a particular topic.
  2. Conversational. The content was written as if addressing the potential reader, e.g., using the second person.
  3. Direct. The scientist applied directly to the audience, e.g., asking for comments or participation. Unlike conversational addresses, direct interaction showed an expectation of response.
  4. Active. The scientist engaged in an active dialogue with the audience, e.g., through comments.

Verbosity is a second dimension for the description of content units. The combination of the two dimensions content type and verbosity allows a complex characterisation of content on scientists’ profiles. Assessing verbosity in addition to content type helps to distinguish among the different ways that scientists use to develop similar contents. For example, scientists can inform about their research activities by listing the areas of their research interest or by discussing in detail their experience from a particular research project.


Each profile instance of an individual scientist was also viewed with regard to the scientist’s other profiling engagements. Two points were found relevant for the placement of the profile instances: placement on a particular type of a platform and placement within the profile network.

The scientists in the sample have created profile instances on a number of different platforms (e.g., institutional or private HTML, SNS, blog, microblog). In order to understand the profiling patterns, the placement of information with regard to the selected platform type was monitored. Furthermore, the profile instances belonging to a scientist’s virtual profile were (partially) connected by hyperlinks. The profile network could thus be depicted as directed graphs. The placement of an instance within a network differed with regard to the needs and the online profiling behaviour of the scientist. Certain profile instances were connected with a visibly higher intensity than others, serving as central points of the network. Some instances attracted a relatively high number of incoming hyperlinks and could be described as ‘information points’. Other instances had a large number of outgoing hyperlinks and thus served in the network as ‘hubs’. The ability of a profile instance to serve as an information point was not limited by the platform, where it was placed. However, if an instance was to serve as a hub, it had to be able to accommodate multiple hyperlinks to different platform types. Some platforms limit the number of hyperlinks or the hyperlink types (e.g., Twitter). Although creative ways were devised to include further hyperlinks (e.g., as a background image on Twitter), overall, such platforms rarely hosted hub instances.

The dimension placement can be used for analysis on the level of profile instances and profile networks. The platform placement of a profile instance and the position in the profile network can be combined with the information about the content of the instance and used to identify profiling patterns.




The analytical framework was constructed as a result of a data–driven analysis of 79 profiles belonging to 15 German scientists. In the following, I discuss how my results relate to existing literature with regard to study focus, researched objects, and identified dimensions.

I have studied four types of platforms that can be used by scientists to create online profiles: static HTML pages (institutional or private), SNS, blogs, and microblogs. Existing studies have focused only on single platform type (e.g., personal homepages, Dillon and Gushrowski, 2000; blogs, Herring, et al., 2005; or SNS, Strufe, 2010). As such, they could discuss the profile information with higher granularity than I have done. However, scientists can have several online profiles (Stewart, et al., 2008). Thus to understand better what information scientists provide about themselves online, I have chosen to adopt a more holistic approach. Consequently, to analyse information across platforms, I had to create abstract categories. Furthermore, unlike most existing studies, the analytical framework focuses directly on scientists’ use of the online platforms and thus cannot be applied to Internet users in general.

The object of this study was the content of scientists’ profiles. Although the profile instances were grouped into cases belonging to particular scientists, the scientists’ behaviour (e.g., reasons, decision) was not analysed. A different approach is adopted by Schmidt (2007), who has developed a framework for the classification of blogs. Schmidt considers blogs as tools for information, identity, and relationship management. The framework is based on the idea of blogging practices, that consist of individual blogging episodes. The focus of the framework is thus on the blog owners and their decisions and choices. Similarly, Hess (2002) studied how and why faculty members manage their Web pages, with content a result of profiling behaviour of a given author. Ferguson, et al. (2010) and Ewins (2005) also study scientific blogging as a form of behaviour, thus placing the blog author in the centre of their study. The analytical framework in this article could be applied in studies focusing on the profile owners, but it would have to be supplemented by further categories and dimensions (e.g., combined with the framework by Schmidt, 2007).

The described analytical dimensions correspond to findings in existing literature. Schmidt (2007) considers blogs as tools for information, identity and relationship management. This is similar to the verbosity categories factual information, identification, and personalisation respectively. Halavais (2006) and Walker (2006), who discuss academic blogs, base their typologies on the types of content. Blood (2002) categorises implicitly by level of personalisation and amount of factual information. The blog typology of Herring, et al. (2005) can be interpreted as using type of content and personalisation. Carroll, et al. (2009) as well as authors concerned with digital identity (Wessels, 2009; Mesch and Talmud, 2006; Strufe, 2010; Bakardjieva and Feenberg, 2000; Satchell, et al., 2006; Grasmuck, et al., 2009) and networking (Wang, et al., 2010; Ferguson, et al., 2010; Kjellberg, 2010; Ebner and Maurer, 2009) point to the importance of social interaction and activity. These are partially mirrored by the verbosity categories personalisation and interaction. Finally, Hargittai (2008) notes that Internet users uses Internet platforms in different ways and warns against generalisation of findings. This points to the importance of the placement dimension of my framework.




In this study, I have analysed the content of scientists’ profiles on different Internet platforms. I have identified three levels of analysis, each providing a different view on the scientists’ online presence: (1) the profile network, as a collection of interlinked profiles on different platforms belonging to one scientist; (2) the profile instance, containing all information provided by a scientist within a profile on one platform; and, (3) the content unit, representing a package of information with similar characteristics within a profile instance. The content units can be analysed with regard to three dimensions: (1) type of content, describing the presented topics; (2) verbosity, assessing qualitatively the amount of content; and, (3) the placement of the information on the platform as well as within the profile network. The combined findings form an analytical framework (see Figure 2).

The results presented here contribute to the area of computer–mediated scientific communication and collaboration. In this area, scholars have profited from the dynamic development of information and communication technologies, particularly the Internet, gaining new possibilities to present themselves and their work. But in order to unleash further potential, it is necessary to understand how the scientists use their online profiles. Then it is possible to address issues like strategic management of online identity, aggregation of profile information, or search for potential collaboration partners. The analytical framework provided in this study can be used for explorative research on scientists’ online presence. The framework takes into account different dimensions and levels of analysis. It can thus serve as a foundation for structured, yet complex, multi–faceted studies.

There are, however, limitations to the findings. Firstly, the study was based on the constructivist grounded theory method, which I acknowledge to rely on subjective construction and interpretation. To address this issue, the results were discussed continually with other scientists (colleagues as well as a broader community), to introduce different understanding and views. Secondly, I have studied a small sample of German scientists who are very active in the Internet. Furthermore, the definition of ‘a scientist’ is limiting. Therefore, the results cannot be readily generalised to the population of scientists present online. However, as the findings show correspondence to existing research results and as the derived dimensions and categories are on an abstract level, the framework should be applicable beyond the sampled cases. End of article


About the author

Helena Bukvova is researcher and a lecturer at the Chair of Information Management at the Dresden University of Technology, Germany. Her research work is concerned with e–collaboration in different settings, such as in e–learning, enterprise 2.0 and scientific communication.
E–mail: helena [dot] bukvova [at] tu-dresden [dot] de



I would like to thank Enrico Lovasz, Hendrik Kalb, and Peter Geißler from the Dresden University of Technology, Germany for active exchange of ideas about my research and help with the preparation of the manuscript.













M. Bakardjieva and A. Feenberg, 2000. “Involving the virtual subject,” Ethics and Information Technology, volume 2, number 4, pp. 233–240.

R. Blood, 2002. The Weblog handbook: Practical advice on creating and maintaining your blog. Cambridge, Mass.: Perseus.

J. Carroll, M. Rosson, U. Farooq and L. Xiao, 2009. “Beyond being aware,” Information and Organization, volume 19, number 3, pp. 162–185.

K. Charmaz, 2006. Constructing grounded theory. Thousand Oaks, Calif.: Sage.

K. Charmaz, 2003. “Qualitative interviewing and grounded theory analysis,” In: J. Holstein and J. Gubrium (editors). Inside interviewing: New lenses, new concerns. Thousand Oaks, Calif.: Sage, pp. 311–330.

K. Charmaz, 2000. “Grounded theory: Objectivist and constructivist methods,” In N. Denzin and Y. Lincoln (editors). Handbook of qualitative research Second edition. Thousand Oaks, Calif.: Sage, pp. 509–535.

A. Dillon and B. Gushrowski, 2000. “Genres and the Web: Is the home page the first digital genre?” Journal of the American Society for Information Science, volume 51, number 2, pp. 202–205.<202::AID-ASI11>3.0.CO;2-R

M. Ebner and H. Maurer, 2009. “Can weblogs and microblogs change traditional scientific Writing?” Future Internet, volume 1, number 1, pp. 47–58.

D. Evans, S. Gosling and A. Carroll, 2008. “What elements of an online social networking profile predict target–rater agreement in personality impressions?” Proceedings of the International Conference on Weblogs and Social Media (Seattle, Wash.), pp. 1–6.

R. Ewins, 2005. “Who are you? Weblogs and academic identity,” E-learning and Digital Media, volume 2, number 4, pp. 368–377.

U. Farooq, C. Ganoe, J. Carroll and C. Giles, 2007. “Supporting distributed scientific collaboration: Implications for designing the CiteSeer collaboratory,” HICSS ’07: Proceedings of the 40th Hawaii International Conference on System Sciences, p. 26c.

R. Ferguson, G. Clough and A. Hosein, 2010. “Shifting themes, shifting roles: The development of research blogs,” ATL–C 2010: Proceedings of the 17th Association for Learning Technology Conference (7–9 September; Nottingham, U.K.), pp. 111–117.

C. Frankfort–Nachmias and D. Nachmias, 2008. Research methods in the social sciences. Seventh edition. New York: Worth Publishers

P. Genoni, H. Merrick and M. Willson, 2005. “The use of the Internet to activate latent ties in scholarly communities,” First Monday, volume 10, number 12, at, accessed 27 September 2011.

B. Glaser and A. Strauss, 1967. The discovery of grounded theory: Strategies for qualitative research. Chicago: Aldine.

S. Grasmuck, J. Martin and S. Zhao, 2009. “Ethno–racial identity displays on Facebook,” Journal of Computer–Mediated Communication, volume 15, number 1, pp. 158–188.

A. Halavais, 2006. “Scholarly blogging: Moving toward the visible college,” In: A. Bruns and J. Jacobs (editors). Uses of blogs. New York: Peter Lang, pp. 117–126.

E. Hargittai, 2008. “Whose space? Differences among users and non–users of social network sites,” Journal of Computer–Mediated Communication, volume 13, number 1, pp. 276–297, and at, accessed 27 September 2011.

S. Herring, 1996. “Linguistic and critical analysis of computer–mediated communication: Some ethical and scholarly considerations,” Information Society, volume 12, number 2, pp. 153–168.

S. Herring, L. Scheidt, E. Wright and S. Bonus, 2005. “Weblogs as a bridging genre,” Information Technology & People, volume 18, number 2, pp. 142–171.

M. Hess, 2002. “A Nomad faculty: English professors negotiate self–representation in university Web space,” Computers and Composition, volume 19, number 2, pp. 171–189.

I. Holloway and L. Todres, 2006. “Grounded theory,” In: K. Garrish and A. Lacey (editors). The research process in nursing. Fifth edition. Oxford: Blackwell, pp. 192–207.

S. King, 1996. “Researching Internet communities: Proposed ethical guidelines for the reporting of results,” Information Society, volume 12, number 2, pp. 119–128.

S. Kjellberg, 2010. “I am a blogging researcher: Motivations for blogging in a scholarly context,” First Monday, volume 15, number 8, at, accessed 27 September 2011.

G. Mesch and I. Talmud, 2006. “The quality of online and offline relationships: The role of multiplexity and duration of social relationships,” Information Society, volume 22, number 3, pp. 137–148.

J. Mills, A. Bonner and K. Francis, 2006. “The development of constructivist grounded theory,” International Journal of Qualitative Methods, volume 5, number 1, pp. 1–10.

R. Mitchell, 1993. Secrecy and fieldwork. Thousand Oaks, Calif.: Sage.

J. Morse, 2007. “Sampling in grounded theory,” In: A. Bryant and K. Charmaz (editors). The sage handbook of grounded theory. Thousand Oaks, Calif.: Sage, pp. 229–244.

C Satchell, G. Shanks, S. Howard and J. Murphy, 2006. “Knowing me knowing you: End user perceptions of digital identity management systems,” Proceedings of the 14th European Conference on Information Systems (Göteborg, Sweden), at, accessed 27 September 2011.

J. Schmidt, 2007. “Blogging practices: An analytical framework,” Journal of Computer–Mediated Communication, volume 12, number 4, at, accessed 27 September 2011.

D. Sonnenwald, 2007. “Scientific collaboration,” Annual Review of Information Science and Technology, volume 41, pp. 643–681.

S. Stern, 2004. “Expressions of identity online: Prominent features and gender differences in adolescents’ World Wide Web home pages,” Journal of Broadcasting & Electronic Media, volume 48, number 2, pp. 218–243.

A. Stewart, E. Diaz–Aviles and W. Nejdl, 2008. “Mining user profiles to support structure and explanation in open social networking,” Proceedings of the International Workshop on Interacting with Multimedia Content in the Social Semantic Web (Koblenz, Germany), pp. 21–30.

T. Strufe, 2010. “Profile popularity in a business–oriented online social network,” SNS ’10: Proceedings of the Third Workshop on Social Network Systems, and at, accessed 27 September 2011.

J. Walker, 2006. “Blogging from inside the ivory tower,” In: A. Bruns and J. Jacobs (editors). Uses of blogs. New York: Peter Lang, pp. 127–138.

X. Wang, T. Jiang and F. Ma, 2010. “Blog-supported scientific communication: An exploratory analysis based on social hyperlinks in a Chinese blog community,” Journal of Information Science, volume 36, number 6, pp. 690–704.

B. Wessels, 2009. “Locating the self in Web 2.0: Explorations in creativity, identity and digital expression,” Observatorio, volume 3, number 3, at, accessed 27 September 2011.

N. Whiteman, 2007. “The establishment, maintenance and destabilisation of fandom: A study of two online communities and an exploration of issues pertaining to Internet research,” Ph.D. dissertation, University of London.


Editorial history

Received 27 May 2011; accepted 27 September 2011.

Creative Commons License
“Scientists online: A framework for the analysis of Internet profiles” by Helena Bukvova is licensed under a Creative Commons Attribution–NonCommercial–NoDerivs 3.0 Unported License.

Scientists online: A framework for the analysis of Internet profiles
by Helena Bukvova.
First Monday, Volume 16, Number 10 - 3 October 2011

A Great Cities Initiative of the University of Illinois at Chicago University Library.

© First Monday, 1995-2018. ISSN 1396-0466.