Tips for article-writing

I have to recommend this little tip sheet for those who wish to publish acacemic papers in the social sciences. I am one of them and often times I found myself facing a dilemma about how to position or frame an idea. If you agree that a paper usually tells a story – then it is an important decision to frame the story, as an idea could be told in a zillion different ways. Usually I have to choose between the “phenomena” and the “theory”. No matter what decision I make it risks losing one part of the audience as a result of the frame. Ezra Zuckerman calls the “row” (phenomena) and “column” (discipline or theory) problem and usually one cannot aim for both.

I tend to go with the “phenomena” approach when I write, because it is the social phenomena that interest me fundementally. I care less about theory – which is supposed to be the truth about how the world works. But usually we see researchers change their theories (often too quickly) to suit the phenomena and empirical evidences they study. I don’t think highly of that approach either. If theories are just means to “interpret” the empirical evidences, then they lose their most important function – to predict the unknown. In that case, we are not doing research any more – we are no more than investigative journlaists who cannot write as well but may have some fancy statistical skills.

So the “row and column” way of looking at article-writing is the central point he made in the tip sheet. As a consequence of the researcher’s decision to go with the “phenomena” or “theory”, other decisions follow. The whole list can be found here.


September 21, 2008 at 5:58 pm 5 comments

Poster tips

At this ICA I will be presenting a poster, for the very first time. Usually poster is a format widely used in scientific disciplines, but social science conferences start to have poster sessions recently to save time, including the ICA and Sunbelt. I think some studies can actually be explicated better by posters rather than by traditional verbal presentations.

Since I am a newbie of this presentation format, I googled about how to make a good poster and found some really interesting articles. The most comprehensive, and funny, guide of scientific poster making could be found here. Although it says “scientific” posters, most of the advice are equally applicable to non-science posters as well.

May 4, 2008 at 9:44 pm Leave a comment

May update

I haven’t been very diligent in updating my research blog recently …. well, because it took much more time than I had thought to write intelligent blog posts. But it seems that I will be doing that more often in the next three months because I will be doing a lot of reading and blogging about them helps to organize thoughts.

Yes, in case you are wondering, I am finally going to take qualifying exam somewhere this September. The preparation, as well as the exam itself, will be a great opportunity to read, re-read, and organize stuff that I care about. I am excited.

Besides reading, I will also be doing ASIMS, which is a fabulous workshop on research methods, hosted by the Annenberg School at USC. Both USC and Penn students will join the two week workshop. I am tentatively in for categorical analysis and multi-level analysis. Sounds like great fun.

I am also presenting at the annual ICA conference at Montreal from May 21-26:

Shen, C. & Monge, P. (2008, May). Power asymmetry and network structure in Open Source community. Paper presented at the 58th Annual Conference of the International Communication Association (ICA), Montreal, Canada.

Shen, C. & Fulk, J. (2008, May). Ecological dynamics of online communities. Paper presented at the 58th Annual Conference of the International Communication Association (ICA), Montreal, Canada.

This year’s ICA has a lot of interesting sessions and panels. Besides, it is also time to reunite with old friends (a lot of them!) and make new ones. Can’t wait! 🙂

May 4, 2008 at 9:11 pm Leave a comment

Caring for your introvert

Thanks my fellow Annenberg introvert Trav for sending me Jonathan Rauch’s article “Caring for your introvert”.  Yes, there is another sad hegemony there in our daily social life.

And there is an interesting cultural element too. When I came to the US I started to notice that this is defnitely not a good place for introverts to live. In China, where people are not supposed to speak a lot in classes, being an introvert is okay, still not too good. But in the US where people are judged by what they speak,  being an introvert is absolutely awful.

October 21, 2007 at 6:47 pm 1 comment

Network evolution and field dynamics

I have read this paper at least 3 times and still couldn’t get all the details of it. Seriously, it is a monster, because this paper is a fabulous example of what the network research paradigm could go in the future: the multi-modal, multiplex, multi-level and longitudinal approach, which all the researchers are calling for but few actually do. Have you read anything like that before?

Powell, W. W., White, D. R., Koput, K., & Owen-Smith, J. (2005). Network Dynamics and Field Evolution: The Growth of Interorganizational Collaboration in the Life Sciences. American Journal of Sociology, 111(5), 1463-1568.

Rationale of the study
• Studies of networks are disconnected with analyses of fields. Most network analysis has taken an individual-level or dyadic focus, while very few researchers have taken a macro perspective to study the evolution of entire networks.
• Most existing studies use cross-sectional data, while few employ longitudinal data to study structural change.
• Existing studies are usually focusing on a single network, while few look at the interaction of multiple networks.

The field
• Biotechnology is a highly dynamic and turbulent field, characterized by a high rate of formation and dissolution of linkages and plenty of entry and exit events.
• Six types of organizations and four types of ties.
• Specifically, this paper maps out the shifting network typology and logics of attachment over time.

Mechanisms of network attachment
• H1.-Network expansion occurs through a process in which the most-connected nodes receive a disproportionate share of new ties (accumulative advantage: preferential attachment, rich-get-richer).
• H2.-Network expansion follows a process in which new partners are chosen on the basis of their similarity to previous partners (homophily: social similarity).
• H3.-Network expansion entails herdlike behavior, with participants matching their choices with the dominant choices of others, either in mutual response to common exogenous pressures or through imitative behavior (follow-the-trend: neo-institutionalism, imitating the leaders).
• H4.-Network expansion reflects a choice of partners that connect to one another through multiple independent paths, which increases reachability and the diversity of actors that are reachable (multiconnectivity: search for novelty and interact with heterogeneous partners).

Method and Results
• A number of sophisticated methods are employed to analyze the data:
o Degree distribution: the probabilities of tie formation across different org types and time points.
o Visualization: evolution of the network typology
o Attachment bias (conditional logit models): the logics of attachment
• Summary of results: in the field of biotechnology, network evolution follows a combination of the four logics. Organizations with diverse portfolios of well-connected collaborators are found in the most cohesive, central positions and have the largest hand in shaping the evolution of the field.

Further Questions
• This empirical study links structural analysis with the evolution of the biotechnology field. It presents and tests hypotheses on attachment bias from various theoretical roots, by employing a set of sophisticated methods. After reading this paper, what do you think might be the implications for our own research, in terms of theory, method, and analysis?
• This study compares and analyses network topologies of different types of organizations over time. This structuralist approach corresponds to one of the two types of network studies (the other is connectionist). It puts more emphasis on network typology, while focuses less on the resource flow between the nodes. What might be the pros and cons of this approach?

October 15, 2007 at 10:55 pm 2 comments

computer-mediated group norms

I know it has been a while since I posted the last entry. School is back in full swing and my time is devoted to three classes, two TA discussion sections, and two regular research group meetings. Apart from that, I am also doing four reseach projects and serve as a project manager for ~10 people working for our NSF project right now. All of the above really excite me, but I feel it is a bit difficult for me to keep up with what are going on in the research community….Therefore I am substituting reading notes for “serious” blog entries, but those reading notes are equally thoughtful.

This week and the past week we discussed CMC and group processes in my 641 class.

Postmes, T., Spears, R. & Lea, M. (2000). The formation of norms in computer-mediated communication. Human Communication Research, 26, 341-371

Motivation of study

  • Previous models of social influences of social construction are inadequate in specifying precisely what exerts influence.
  • S ocial construction is a dynamic reciprocal process, which is not accurately captured by rather static operationalizations.

A social identity approach

  • Is context-sensitive, because different situations make different social identities salient

  • SIDE model: mediated group can develop and follow group norms even if they are visually anonymous.

  • Social identity is constantly negotiated through group interaction. This approach allows for the observation of the emergence and development of group norms.

Group norms in CMC setting

  • CMC groups will have more consistencies of interaction content and form within groups than between groups. (Supported).
  • Group norms (in terms of interaction content and form) will be amplified and accentuated over time. (Partially supported for interaction content, but not form).
  • Group norms will influence communication within the group but not outside of the group. (Supported).

My critique

  • Groups were defined as clusters of participants with minimal distance (reciprocal message) in a distance matrix, not based on interaction content/style. Although it is shown that there are significant differences of interaction content and style between groups, the researchers didn’t report within-group differences (individual interaction content and styles) and neither did they compare them with between-group differences.
  • Nothing is known about their prior interaction history. Social norms may have been migrated from other groups formed earlier than this stats course, which could severely undermine the validity of the findings. This actually could explain why interaction style was not amplified over time.
  • The last hypothesis test was not very convincing. Communicating with staff (outgroup) may not necessarily be considered as changing the social context at all. Yes the audience is different, but the social context and group identity may remain the same. It would be interesting to compare people’s interaction content/style with the staff among the same group members.

Discussion questions

  • Researchers didn’t find that interaction styles were amplified over time through group processes. To what extent could individuals develop interaction styles and maintain them in group processes? As we spend more and more time using CMC, how do group processes interact with our individual “style”?
  • (General question) From this article and other CMC research to date, what are the different perspectives to understand the group processes in CMC? How do we reconcile the differences of research findings? What factors and boundary conditions should be considered?
  • (General question) what are the differences between CMC and FtF? When is CMC more/less beneficial than FtF?
  • Considering ecological validity, what are the general methodological challenges of CMC research? What method or combination of methods is good to study CMC?
  • Finally, with the advent of new communication technology and people’s growing comfort of using them, is it accurate to refer to CMC as if it is unitary?


September 30, 2007 at 8:24 pm Leave a comment

Networking for introverts

I study (social) networks, not networking, though people often mistakenly think I should be an expert of networking. So wrong.
I found this article on Businesspundit very helpful as well as the video on “how to not  suck on networking”

Interestingly, it suggests that one’s personality traits, especially extroversion/introversion, are associated with networking ability and networking results, . A borderline extrovert/introvert as I am, I do believe that there are some gifted connectors in the world, while networking ability is also something that begs constant cultivation. One thing I did not realize soon enough about gradschool is that I cannot treat myself as a “real” student, which I often do, but a professional that needs to actively thinking about and planning for her career path.

September 10, 2007 at 12:38 am 2 comments

Older Posts

Time flies

January 2019
« Sep    

Recent Posts

Recent Comments


  • 9,483 hits
Add to Technorati Favorites

Add to netvibes

Subscribe in Bloglines