miércoles, 31 de julio de 2013

Visualización: La red de comercio global

Mapping Globalization: Visualizing the Network of Global Trade

Manish Nag
Doctoral Candidate at Princeton University

Mapping Globalization: Visualizing the Network of Global Trade

Caption
How global is globalization?  The last 20 years have witnessed an explosion of international connections and transactions: we travel more to each other’s cities, buy more of each others’ products, and are more likely to read each others’ newspapers and best-sellers.  But there are severe limits on the reach and degree of globalization.  European and North American newspapers are more likely to be read by an international audience than their Indian or Brazilian counterparts.  Style and sophistication are still more associated with Paris than Shanghai, not to speak of Mumbai.  When we speak the global language of business, science, and the arts, it sounds remarkably like English.
                Even in the most globalized of arenas, the pattern of relations and connections can still look remarkably like the 19th century.  Despite the prophesied rise of a “Pacific Century”, most of the global action remains rooted in the North Atlantic.  While the past decades have witnessed the rise of East Asia, the Global South plays a small role in global trade.  This is particularly true of commerce in manufactures and other high value-added products whose production remains concentrated in relatively wealthy countries.  (The design and sales of these products, where the greatest profit can be derived, is even more concentrated).  When the economies at the global margin do participate, they often do so through the sale of a single commodity or through the export of labor.
                Our animation documents this by tracking global trade in 2001.  Counting all possible dyadic transactions we have taken graphic photographs of four levels of trade. The first image of all global trade includes the numerous, but often insignificant, links connecting the globe.  In the next image of the top 75% of trade, the number of countries involved and lines linking these are noticeably reduced (and the overall geographical concentration and centrality of the United States and Western Europe becomes clear). The next picture subtracts even further to the few commercial relationships needed to account for 50% of the total.  In the last image we can see the relatively few country pairs with the largest commercial transactions adding up to 25% of the global total.

Self-Commentary
                Our visualization was created using Sonoma, a new software tool that exists to create geospatial visualizations of social networks. The tool allows researchers to build maps, and then to automatically overlay social network graphs. As a result of using this tool, creating the visualizations was simply a matter of using Sonoma's user interface to define a map projection and map colors, to upload trade network data in a matrix file, and to upload a separate file for latitude and longitude data for each actor on the map.
                For our example, since nations were our actors, we positioned each graph vertex on the nation's capital city. The latitude and longitude data was obtained from the CIA World Factbook. Once, the data files were uploaded, the Sonoma user interface was used to define visual attributes of the network graph's vertices and ties, along with schemes for scaling the colors and widths of ties based on tie weights.
                The matrix data was furnished by the Mapping Globalization website at Princeton. Though the original data provided directed matrix data for world trade, we converted this data to an undirected format by simply taking the sum of trade in both directions between each dyad of nations. The choice was made to use undirected network data because introducing directional arrows in a global map would create too much visual clutter.
                Once images were created in Sonoma, an animation was rendered using Adobe Photoshop.  Due to the existence of Sonoma, the real challenges in creating the visualization were more in the conception and visual design of the visualization.
Though we created visualizations for other percentages of world trade, we found that choosing the top 25, 50, 75, 100 percentages of world trade summarized the larger point of how much the network of nations shrank as we visualized smaller slices of world trade.

PEER REVIEW COMMENT No. 1
This is a creative animation; when all of the global trade is included it does appear as if global trade is truly global, as the map is literally filled with connections.  But the story is quite different, when one only considers the top 75%, 50% or 25% of the global trade: here the marginal countries drop out of the network and only the major industrialized nations remain.  This large scale visualization tells a clear story in a creative manner, but the figure could perhaps be improved by adding a bit of color to liven the picture or layer more information (such as content of trade, say).  It might also be useful to make the edges more transparent, so that the map shows through even when full.

PEER REVIEW COMMENT No. 2
This visualization uses an interactive layout to show how regions of the world are integrated through trade.  At the most integrated level when 100% of global trade is depicted, the entire world appears integrated. When that level is dropped to 75% of global trade, the picture is very different. The wealthiest nations, and within them – regions, remain. This visualization is very effective already, but perhaps a heat color pattern on the underlying picture or variable line thickness would be a nice addition, to help contextualize each ‘slice.’  The dynamic elements are rhetorically effective – the inequality jumps out in the contrast between the slices – but I wonder how effective it would be to shade ties by proportion of world trade and then layer the information as a single figure?

PEER REVIEW COMMENT No. 3
This map does making a striking clear visual case for the inequality among national actors involved in the global economy.  Its use of edge thresholds leads us naturally to the author’s conclusion without needing to convince us with captions and supplementary material.  I would love to see the edges draw with edge opacity proportional to the trade volume represented by the tie (this might yield a single image that displays all ties, but still permits those few, elite, high-volume ties to stand out).

ARS: El grafo riverplatense

La red social de River, según algunos

Un email anda divulgando una serie de enlaces (como proxy de asociaciones de algún tipo) sobre la red social de las entidades y actores de la escena política del club River Plate (Argentina). ¿Es una red de uno o dos modos? Más allá de la falta de precisiones técnicas, es una atractiva forma de acercar el ARS al escenario de la propaganda política.

domingo, 28 de julio de 2013

vom Lehn: Respuesta a Christakis

christakis vs. dirk vom lehn


Dirk vom Lehn is a lecturer in the Department of Management at King’s College London. His research focuses on ethnomethodology in organizational settings. He asked if I could post this response to Christakis’ NY Times article on the need to update the social sciences.
Stagnating the Social Sciences? A response to Nicholas Christakis?
In his recent piece “Let’s Shake Up the Social Sciences” published in the New York Times on July 19th, Nicholas Christakis calls for interdisciplinary research that creatively links the social sciences to other disciplines, in particular the natural sciences. I very much welcome his efforts to open a debate about the future of the social sciences. All too often scientists create separate enclaves of knowledge that, if joint up with others, could lead to important new academic, technological and political developments. There however are a few problems with Christakis’ argument. I wish to briefly address three of these problems here:
I am surprised Christakis puts forward the argument that “the social sciences have stagnated” over the past years. He gives no empirical evidence for such a stagnation of the social scientific disciplines and I wonder what the basis for this argument is. If he was to attend the Annual Conference of the American Sociological Association (ASA) in New York in August he will see how sociology has changed over the past few decades, and he will be able to identify specific areas where sociologists have impacted developments in policy, technology, medicine, the sciences, the arts and elsewhere.
His argument ignores also the long-standing cooperation between social scientists, technology developers, computer scientists, medics and health services providers, policy makers, etc. etc. etc. For example, for several decades social scientists, computer scientists and engineers have collaborated at research labs of PARCs,  Microsoft and elsewhere, jointly working to develop new products and services.
Christakis refers to the development of new fields like neuroscience, behavioral economics and others that “lie at the intersection of natural and social sciences”. Because “behavioral economics” is popular also with policy makers let us take this new field as an example: one of the key findings of this new field is the importance of “non-rational action” for people’s decision making. I very much enjoy the creative research undertaken by scholars in this field, but it is quite surprising that it gets away with by-and-large disregarding 100 years of social scientific research. Critique of arguments that prioritize rational action over other types of action has been key to Max Weber’s famous work in the early 1900s, Talcott Parsons’ discussion of the utilitarian dilemma, Harold Garfinkel’s breaching experiments and many other sociologists’ research and teaching.
Speaking of Garfinkel and his breaching experiments: Christakis suggests that social scientists do not use lab experiments in their teaching. He might be pointed to Garfinkel who used experiments or “tutorial exercises”, as he called them, on a regular basis to have students discover how people organize their action and interaction that bring about society. Experimental research has been conducted also by Carl Couch and the Iowa School since the 1960s with the aim to identify the key elements of social relationships. And, there are a considerable number of more social scientists who have used lab experiments to understand social action and interaction.
However, it has been noticed since that time that society does not happen in the lab. Therefore, in many social scientific disciplines lab experiments are rarely seen as the best way forward to find out about the organization of society. Garfinkel, for example, has continued to use tutorial demonstrations in his teaching but increasingly looked into the organization of the everyday world as it manifested itself in waiting queues, traffic jams and elsewhere. And the Iowa School and its experimental approach has largely vanished whereby its methods and findings can be found in symbolic interactionism and other areas. While the influence of experimental approaches has diminished, naturalistic, ethnographic and video-based research has come to the fore, most notably in workplace studies, in studies of interaction in urban environments and public places as well as in online environments. This body of studies builds on a history of more than 100 years of sociological ethnography, going back, for example, to Robert Park, Everett Hughes and the Chicago School of Sociology. Here, sending students into the field, i.e. into workplaces and schools, onto city streets, on street-markets, into museums, into parks, into Second Life and other virtual worlds, etc. has been at the center of education, training and research as it allows students to discover first-hand how society works. Scholars also increasingly use video-based research to explore the practical organization of work in complex organizations, such as operating theaters in hospitals, control rooms of rapid urban transport systems, museums and galleries, etc.
Christakis’ article is an unfortunate case of a contribution to a debate that means well in steering up discussion about the future of the social sciences, that however ends up playing into the hands of those who have launched an “attack on the social sciences”, as Sally Hillman, Executive Officer at the American Sociological Society, has called it in the association’s newsletter ‘footnote’ in June. Senators and members of the House Science Committee have suggested to “defund” Political Science at the NSF and proposed bills that “would […] prevent NSF from funding any social science research” (Hillman June 2013).
Articles like Christakis’ imply that current social sciences have little impact on society, policy makers and knowledge development more generally, whilst research in the natural sciences, in their view, has more “impact”. They, however, overlook and disregard social scientific research that has been forgotten because scholars and policy makers follow the latest fads and fashions, such as so-called Big Data research and the opportunities of brain-scans, rather than using and further developing the existing theoretical, methodological and empirical basis of the social sciences. Moreover, they pretend that the social sciences and the natural sciences basically could achieve the same impact, if only the social sciences would make appropriate use of scientific methods. Thereby, however, they ignore what social scientists have shown over and over again over the past 100 years or so, i.e. that the social is fundamentally different from nature; it always is already interpreted when the social scientist arrives. The ‘social’ requires interpretation of a different kind than nature as encountered and then interpreted by natural scientists. Furthermore, people often change their behavior in response to the research process and in response to social scientific findings. Nature remains nature. Apples keep falling down from trees.
I am all in favor of interdisciplinary research and benefit enormously from my cooperation with scholars and practitioners in the computer and health sciences as well as in the arts and humanities. I also find Christiakis’ research interesting and important. However, to use the need for interdisciplinarity as an argument for the defunding of established social science disciplines would be like throwing the baby out with the bath water. The social scientific knowledge base developed over the past 100 or more years is too precious to sacrifice just for instrumental reasons; i.e. to satisfy policy makers interested in saving money or to show “impact” however that is defined.
While the social sciences rely on and advance their knowledge base they have not been stagnating. On the contrary, they have prospered and further developed by virtue of discussions at discipline-specific conferences and in their journals as well as by cooperating with a wide range of other disciplines.
Dr Dirk vom Lehn
Lecturer in Marketing, Interaction & Technology
Department of Management
King’s College London
Franklin-Wilkins Building, 150 Stamford Street
London SE1 9NH
Tel. +44 20 78484314
dirk.vom_lehn@kcl.ac.uk

Christakis: Un sacudón hacia la complejidad en las ciencias sociales

Let’s Shake Up the Social Sciences




TWENTY-FIVE years ago, when I was a graduate student, there were departments of natural science that no longer exist today. Departments of anatomy, histology, biochemistry and physiology have disappeared, replaced by innovative departments of stem-cell biology, systems biology, neurobiology and molecular biophysics. Taking a page from Darwin, the natural sciences are evolving with the times. The perfection of cloning techniques gave rise to stem-cell biology; advances in computer science contributed to systems biology. Whole new fields of inquiry, as well as university departments and majors, owe their existence to fresh discoveries and novel tools.

In contrast, the social sciences have stagnated. They offer essentially the same set of academic departments and disciplines that they have for nearly 100 years: sociology, economics, anthropology, psychology and political science. This is not only boring but also counterproductive, constraining engagement with the scientific cutting edge and stifling the creation of new and useful knowledge. Such inertia reflects an unnecessary insecurity and conservatism, and helps explain why the social sciences don’t enjoy the same prestige as the natural sciences.
One reason citizens, politicians and university donors sometimes lack confidence in the social sciences is that social scientists too often miss the chance to declare victory and move on to new frontiers. Like natural scientists, they should be able to say, “We have figured this topic out to a reasonable degree of certainty, and we are now moving our attention to more exciting areas.” But they do not.
I’m not suggesting that social scientists stop teaching and investigating classic topics like monopoly power, racial profiling and health inequality. But everyone knows that monopoly power is bad for markets, that people are racially biased and that illness is unequally distributed by social class. There are diminishing returns from the continuing study of many such topics. And repeatedly observing these phenomena does not help us fix them.
So social scientists should devote a small palace guard to settled subjects and redeploy most of their forces to new fields like social neuroscience, behavioral economics, evolutionary psychology and social epigenetics, most of which, not coincidentally, lie at the intersection of the natural and social sciences. Behavioral economics, for example, has used psychology to radically reshape classical economics.
Such interdisciplinary efforts are also generating practical insights about fundamental problems like chronic illness, energy conservation, pandemic disease, intergenerational poverty and market panics. For example, a better understanding of the structure and function of human social networks is helping us understand which individuals within social systems have an outsize impact when it comes to the spread of germs or the spread of ideas. As a result, we now have at our disposal new ways to accelerate the adoption of desirable practices as diverse as vaccination in rural villages and seat-belt use among urban schoolchildren.
It is time to create new social science departments that reflect the breadth and complexity of the problems we face as well as the novelty of 21st-century science. These would include departments of biosocial science, network science, neuroeconomics, behavioral genetics and computational social science. Eventually, these departments would themselves be dismantled or transmuted as science continues to advance.
Some recent examples offer a glimpse of the potential. At Yale, the Jackson Institute for Global Affairs applies diverse social sciences to the study of international issues and offers a new major. At Harvard, the sub-discipline of physical anthropology, which increasingly relies on modern genetics, was hived off the anthropology department to make the department of human evolutionary biology. Still, such efforts are generally more like herds splitting up than like new species emerging. We have not yet changed the basic DNA of the social sciences. Failure to do so might even result in having the natural sciences co-opt topics rightly and beneficially in the purview of the social sciences.
New social science departments could also help to better train students by engaging in new types of pedagogy. For example, in the natural sciences, even college freshmen do laboratory experiments. Why is this rare in the social sciences? When students learn about social phenomena, why don’t they go to the lab to examine them — how markets reach equilibrium, how people cooperate, how social ties are formed? Newly invented tools make this feasible. It is now possible to use the Internet to enlist thousands of people to participate in randomized experiments. This seems radical only because our current social science departments weren’t organized to teach this way.
For the past century, people have looked to the physical and biological sciences to solve important problems. The social sciences offer equal promise for improving human welfare; our lives can be greatly improved through a deeper understanding of individual and collective behavior. But to realize this promise, the social sciences, like the natural sciences, need to match their institutional structures to today’s intellectual challenges.


Nicholas A. Christakis, a physician and sociologist at Yale University, is a co-director of the Yale Institute for Network Science.

sábado, 20 de julio de 2013

Los jovénes ricos más proclives a conductas de levante

Why Wealthy College Kids Are More Into 'Hooking Up' Than Their Poorer Classmates




The New York Times' recent article on the hookup culture at the University of Pennsylvania has garnered many fiery responses, with current and former Penn students blasting the piece for everything from its casual treatment of rape to its reliance on anonymity


Even the Daily Pennsylvania — UPenn's student newspaper — has come out swinging against the Times, noting:
The response has been largely critical, and for good reason: In her failed attempt to glimpse a part of Penn’s culture, Taylor drew conclusions that inaccurately represented and overly generalized the University’s student body.
Despite its many issues, one tidbit deeply buried in the article deserves more attention.
According to the article, two researchers who followed a group of women at IndianaUniversity from their freshman to senior years found that "women from wealthier backgrounds were much more likely to hook up, more interested in postponing adult responsibilities and warier of serious romantic commitment than their less-affluent classmates. The women from less-privileged backgrounds looked at their classmates who got drunk and hooked up as immature."
In an excellent breakdown of the Times' article on Jezebel, writer Tracy Moore sums this up as "Rich ladies like to PARTY, poor girls have to keep it real...virginal." Moore notes that she finds class-based attitudes towards casual sex "actually really fascinating," although "it's almost never explored."
While the research on privilege and hooking up is sparse at best, a 2009 study from Laura Hamilton and Elizabeth A. Armstrong — the two researchers quoted in the Times — sheds some light on this often overlooked field. After speaking with a diverse group of women at a Midwest university, here's what they found:
  • The dominant college culture "reflects the beliefs of the more privileged classes."
  • Less privileged women's disinterest in hooking up could be "characterized by a faster transition into adulthood."
  • For less privileged women not interested in hooking up, "college life could be experienced as mystifying, uncomfortable, and alienating."
  • 40% of less privileged women left the university, compared to 5% of more privileged women, and "In all cases, mismatch between the sexual culture of women’s hometowns and that of college was a factor in the decision to leave."
  • Many less privileged women left the university because of pressure from boyfriends or husbands back home, or to end hometown speculation that they were now a "slut."
  • Because less privileged women "had less exposure to the notion that the college years should be set aside solely for educational and career development"they often did not view "serious relationships as incompatible with college life."
While the entire report is fascinating and worth a read, its findings are not without opponents. Although not a direct response to Hamilton and Armstrong's paper, a 2012 study on college sexual activity from researchers at Georgia Southern University found that social class is not a "significant predictor of hooking up." Clearly, this is an area that still requires some study.

viernes, 19 de julio de 2013

Los hombres gastan más mensajes en hacer contacto

Here's How Many Messages Men Have To Send To Women On A Dating Site To Be Sure Of Getting A Response

Business Insider

Yesterday, we posted a chart that Josh Fischer at Snap Interactive (STVI: OTC BB) sent us based on analytics from their dating website Are You Interested. 
It shows the likelihood that a someone on AYI.com responds to a message from a member of the opposite sex given their age difference. To the left, -10 means the sender was 10 years younger, on the right, the sender was 10 years older, with zero indicating that the sender and the recipient are the same age. 
Here are the two plots for men responding to messages from women (blue) and women responding to messages from men (red).

Josh Fischer / SNAP Interactive
So, we can see that women are much more selective than men when it comes to responding to messages. Not exactly Nobel-quality findings here, but it's definitely interesting to see the exact levels of response.
But for folks in the dating game, how is this information actually usable?
Well, let's find out how many messages the average man will have to send to a woman his own age in order to guarantee various levels of response, and vice versa.
We can't guarantee a response, per se, but we can say how confident we are that these average bachelors and bachelorettes will receive at least one response given the number of messages they send.
We know, from the chart above, that a woman who sends a message to a man her own age has a 17.5% likelihood of receiving a response to that message.
We know that a man who sends a message to a woman his own age has a 4% likelihood of receiving a response to that message. 
Extrapolating from there, here's how confident men and women can be that they will receive a response given the number of messages they send en masse:

Walter Hickey / BI
Fascinating. 
An average man who sends 18 messages to women his own age can be 50% certain he'll receive at least one response. For women, they need to send only 5 messages to be 50% certain they'll get a response. 
Looking at higher confidence levels, if a woman wants to be 90% certain she'll receive a response from a man her own age, she'll have to send 13 messages. A man will have to send 58 messages.
Finally, to be 99% certain she'll receive a response, a woman must send 25 messages to men her own age. 
A man will have to send 114 .
This leads us to believe that one potential cause of the disparity between the male response rate and female response rate is the system itself.
If men must spam women with messages in order to elicit a response, then women will be more selective when responding to the surplus in general. Since women are understandably disinclined to respond to all the messages, men must send out more in order to guarantee any response. It's cyclic. 
Everyone is acting in their own self interest, inadvertently leading to further imbalances in the system. 
It's one of the fundamental issues with online dating in general. 
Anyway, best of luck out there folks. 

jueves, 18 de julio de 2013

Cómo Node XL ayuda a investigar las comunidades de Twitter

Look Out Klout, These Twitter Influencer Maps Are Amazing

Mark Fidelman, Contributor
Forbes
What if instead of a score, you could visualize the impact a person, business or topic has in a social network? What if instead of using complicated listening tools, you could see in an instant who is talking about your company or its products and how you’re connected to them? What if you could tell who the major influencers or connectors are that everyone else is listening to? From what I’ve seen from the social network maps from NodeXL, this is all possible and a whole lot more.
It’s difficult to have a discussion today with any organization who doesn’t tell you some version of this: Advertising is becoming less effective, yet I can’t figure out why, let alone understand how to adapt. I keep hearing that advocate and influencer marketing is more effective, but I haven’t figured out how to find the right people that clearly impact my potential buyer.
Well allow me to speculate. We now live in an increasingly saturated world — a world where ads are placed nearly everywhere – a world with so much marketing noise it’s hard to separate fact from fiction – a world where even the big TV networks and news corporations no longer have the ability to create overnight product successes.
Our saturated world is causing people to tune out of traditional forms of advertising and turn to industry experts or influencers to make sense of it for them. The old magic formula for selling product through proven media channels has disappeared. So smart organizations are turning to thought leaders, influencers and experts to endorse and promote their products.
But how do you accurately find them and how do you know who is really having an impact on the organization? Moreover, can I look at the influencer or expert’s network to understand who they influence?
Take a look at my NodeXL Twitter Social Graph given to me by Social Network Theory expert Marc Smith:
To the untrained eye, the map is a little difficult to decipher (Smith is working on simplifying it for the rest of us) but I am at the center of that blue mass in section G1. Smith tells me that the Forbes community in G3 is supportive of my articles (that’s good to know) but there are a whole lot of people in G2 that I don’t know and are not connected with that are discussing topics related to me. Smith says that I should make the effort to get to know them because they could be very helpful with my goals. In G4 – G34, those are communities of people that I may or may not know – but are discussing my content – and I should seek to extend my network into those communities.

The Six Basic Types of Twitter Social Networks

Below using analysis from NodeXL, Smith outlines the 6 major types of Twitter social network types. What’s exciting is that this tool empowers companies to understand not only the social network around them, but who is most important in it. This huge expansion in social network understanding allows organizations to find hidden prospective customers, connect with the true influencers in their industry, and A/B test the impact of their social campaigns (NodeXL allows multiple ways of looking at the network data).
Let’s briefly explore each type, and then I’ll provide some relevant, real world examples for you to look at on the NodeXL site:
1. Polarized Network: Most often seen in politics or political issues, this pattern emerges when two groups are split in their opinion on an issue. Here’s one on Egyptian crisis.
2. In-Group Network: Seen at conferences and tight knit groups of people, this type of network rarely ventures outside of its membership. A big miss in most cases if you are a brand. Here’s an example of one for Social Business.
3. Brand/Public Topic Network: Most often seen when a person or company becomes a brand and people other than your customers are talking about you. Notice all of the disconnected users that are isolated from the brand and not connected. Car companies are a good example of a brand network.
4. Bazaar Network: Most often seen with medium sized companies or political issues with various community involvement (see the Texas Abortion Law example)
5. Broadcast Network: As its name implies, these individuals or companies have the power to light up the network – or in this baseball team example, have the ability to light up the Twitter scoreboard. News organizations also display a classic broadcast pattern.
6. Support Network: Think customer support. These types of networks are known to be good at customer service.

In case you’re wondering, a lot of this research came out of Microsoft or in partnership with Microsoft research. The following are other people and institutions that are involved including Stanford, The University of Washington and the University of Maryland.
clip_image005
Truly, these maps are very impressive and a big leap for organizations wishing to understand their impact in social networks. But for companies and individuals to take advantage of the opportunities presented in the network maps, NodeXL will need to make it easier for users to understand them and take action. Smith assures me this is right around the corner.
In fact, Smith tells me the maps will be much more interactive and actionable. That will make finding influencers and new customers incredibly easy.
That’s a big game changer.

lunes, 15 de julio de 2013

8 herramientos analíticas de Twitter

8 Excellent Twitter Analytics Tools to Extract Insights from Twitter Streams

Yung Hui Lim


Twitter is now the third most popular social network, behind Facebook and MySpace (Compete, 2009). A year ago, it has over a million users and 200,000 active monthly users sending over 3 million updates per day (TechCrunch, 2008). Those figures have almost certainly increased since then. With the torrential streams of Twitter updates (or tweets), there's an emerging demand to sieve signals from noises and harvest useful information.
Enter Twitter Analytics, Twitter Analysis, or simply just Analytwits (in the tradition of Twitter slang). These analytics tools are growing in numbers; even Twitter is developing them.
Besides Twitter Search, the following 8 Analytwits are some of the more useful web applications to analyze Twitter streams. Each of these tools serve specific purpose. They crawl and sift through Twitter streams; also, aggregate, rank and slice-and-dice data to deliver some insights on Twitter activities and trends. There's no single best analytic tool available but use in combination, they can extract interesting insights from Twitter streams.
8 Great Tools for Social (Twit)telligence

twitalyzerTWITALYZER provides activities analysis of any Twitter user, based on social media success yardsticks. Its metrics include (a) Influence score, which is basically your popularity score on Twitter (b) signal-to-noise ratio (c) one's propensity to ‘retweet' or pass along others' tweets (d) velocity - the rate one's updates on Twitter and (e) clout - based on how many times one is cited in tweets. Its Time-based Analysis of Twitter Usage produces graphical representation of progression on various measures. Using Twitalyzer is a easy; just enter your Twitter ID and that's it! It doesn't require any password to use its service. Speed of analysis is depending on the size of your Followed and Followers lists.

microplazaMICROPLAZA offers an interesting way to make sense of your Twitter streams. Called itself “your personal micro-news agency,” it aggregates and organizes links shared by those you follow on Twitter and display them as newstream. Status updates that contain similar web links are aggregated into 'tiles.' Within a tile, you can see updates from those you follow and also those you don't. Another interesting feature is ‘Being Someone', which you can peek into someone's world and see their 'tiles'; designed to facilitate information discovery. You can also organize those you follow into groups or ‘tribes'. You can create, for example, a knitting ‘tribe' to easily what URLs your knitting friends are tweeting. In addition, you can bookmark 'tiles' for future reference. Its yet-to-be-released feature, Mosaic, allows users to group together the bookmarked 'tiles' and turn them into social objects - for sharing and discussion. At the time of this posting, MicroPlaza is still in private beta.

twistTWIST offers trends of keywords or product name, based what Twitter users are tweeting about. You can see frequency of a keyword or product name being mentioned over a period a week or a month and display them on a graph. Select an area on the graph to zoom into trend for specific time range. Click on any point on the graph to see all tweets posted during a specific time. One can also see the latest tweets on the topic. Twist also allows you do a trend comparison of two (or more) keywords. Its graphs are embeddable on any website. A simple but effective tool for trending, similar to what Google Trends is doing for search queries.

TwitturlyTWITTURLY tracks popular URLs tracker on Twitter. With Digg-style interface, it displays 100 most popular URLs shared on Twitter over the last 24 hours. On Digg, people vote for a particular web content, whereas on Twitterurly, each time a user share a link, it is counted as 1 vote. This is a good tool to see what people are ‘talking' about in Twitterville and see total tweets that carry the links. Its URL stats provides information on number of tweets in last 24 hrs, last 1 week and last 1 month. It also calculates total estimated reach of the tweets. Another interesting site is Tweetmeme, which can filter popular URLs into blogs, images, videos and audios.

TweetStatsTWEETSTATS is useful to reveal tweeting behavior of any Twitter users. It consolidates and collates Twitter activity data and present them in colorful graphs. Its Tweet Timeline is probably the most interesting, as it shows month-by-month total tweets since your joined Twitter (TweetStats showed Evan Williams, co-founder of Twitter, started tweeting since March 2006; 80 tweets during that month). Twitterholic can also show when a person joined Twitter but not in graphical format. Other metrics include (a) Aggregate Daily Tweets - total tweets, by day (c) Aggregate Hourly Tweets - total tweets, by hour (d) Tweet Density: hourly Twitter activities over 7 days period (e) Replies to: top 10 persons you've replied and (f) Interfaces Used: top 10 clients used to access Twitter. In addition, its Tweet Cloud allows you to see the popular words you used in your tweets.

TwitterFriendsTWITTERFRIENDS focuses on conversation and information aspects of Twitter users' behaviors. Two key metrics are Conversational Quotient (CQ) and Links Quotient (LQ). CQ measures how many tweets were replied whereas LQ measures how many tweets contained links. Its TwitGraph displays six metrics - Twitter rank, CQ, LQ, Retweet Quotient, Follow cost, Fans and @replies. Its interactive graph (using Google Visualization API) can displays relationships between two variables. In addition, you can search for conversations between two Twitter users. This app seems to slice-and-dice data in more ways compared to other applications listed here.

ThummitTHUMMIT QUICKRATE offers sentiments analysis, based on conversations on Twitter. This web application identifies latest buzzwords, actors, movies, brands, products, etc. (called ‘topics') and combines them with conversations from Twitter. It does sentiment analysis to determine whether each Twitter update is Thumms up(positive), neutral or Thumms down (negative). Click on any topic to display opinions on the topic found on Twitter. In addition, it allows people to vote on topics via its website or mobile phones. The idea behind this app is good but still has some kinks to work out.

TweetEffectTWEETEFFECT matches your tweets timeline with your gain/lose followers timeline to determine which tweet makes you lost or gain followers. It analyze the latest 200 tweets and highlights tweets that coincides with you losing or gaining two (or more) followers in less than 5 minutes. This application simplistically assumed that your tweet is the sole factor affecting your gain/lose followers pattern. But, in reality, there are many other factors involved. Nevertheless, TweetEffect is still a fun tool to use; just don't take the results too seriously.

Let's Continue the Discourse on Twitter
Which of the abovementioned Twitter analytics you like the most? How can these tools generate revenue? Have you discovered any other interesting Twitter analytics? Share your thoughts on Twitter; find me @limyh

viernes, 12 de julio de 2013

Factor de impacto en los journales: ¿Para que sirven?

Journal impact factors: what are they good for?



The ISI journal impact factors for 2012 were released last month. Apparently 66 journals were banned from the list for trying to manipulate (through self-citations and “citation stacking”) their impact factors.

There’s a heated debate going on about impact factors: their meaning, use and mis-use, etc.  Science has an editorial discussing impact factor distortions.  One academic association, the American Society for Cell Biology, has put together a declaration (with 8500+ signers so far)–San Francisco Declaration on Research Assessment (DORA)–highlighting the problems caused by the abuse of journal impact factors and related measures. Problems with impact factors have in turn led to alternative metrics, for example see altmetrics.
I don’t really have problems with impact factors, per se.  They are one, among many, measures that might be used to measure journal quality.  Yes, I think some journals indeed  are better than others.  But using impact factors to somehow assess individual researchers can quickly lead to problems.  And, it is important to recognize that impact factors assume that articles within the journal are homogeneous, though within-journal citations of course are radically skewed.  Thus a few highly-cited pieces essentially prop up the vast majority of articles in any given journal. Citations might be a better measure, though also highly imperfect.  If you want to assess research quality: read the article itself.
On the whole, article effects trump journal effects (as Joel Baum’s article also points out, see here).  After all, we all have one-two+ favorite articles, published in some obscure journal no one has ever heard of.  Just do interesting work and send it to journals that you read.  OK, that’s a bit glib.  I know that all kinds of big issues hang in the balance when trying to assess and categorize research: tenure and promotion, resource flows, etc. Assessment and categorization is inevitable.
A focus on impact factors and related metrics can quickly lead to tiresome discussions about which journal is best, is that one better than this, what are the “A” journals, etc.  Boring.  I presented at a few universities in the UK a few years ago (the UK had just gone through its Research Assessment Exercise), and it seemed that many of my interactions with young scholars devolved into discussing which journal is an “A” versus “A-” versus “B.”  Our lunch conversations weren’t about ideas – it was disappointing, though also quite understandable since young scholars of course want to succeed in their careers.
Hopefully enlightened departments and schools will avoid the above traps and focus on the research itself.  I think the problems of impact factors are well-known by now and hopefully these types of metrics are used sparingly in any form of evaluation, and only as one imprecise datapoint among many others.
[Thanks for Joel Baum (U of Toronto) for sending me some of the above links.]

martes, 9 de julio de 2013

El Facebook del Senado estadounidense

The Senate as Facebook



Ever wonder what the Senate would look like viewed through the lens of Facebook?  Us too.


This is Facebook.
Now, thanks to Yahoo’s Chris Wilson, we know. Using Senate votes, Wilson has created a mini-social network of the world’s greatest deliberative body.
“For every member, I calculated which other senators voted the same way at least 75 percent of the time. In effect, this organizes the Senate as a mini-Facebook of 100 users, in which any given pair of senators are friends if they meet this 75-percent threshold….Visualizations like this one work by treating the senators as particles that repel one another, and treating the connections between them as springs that hold them together. Because the Democrats vote so cohesively, with few defectors, they are held together by a large number of springs.”
In the chart below, you can see the Senate as a whole or sort via specific Senator to see whether they have any ties — meaning they vote with a colleague 75 percent or more of the time — to other Senators.
What’s clear from the chart is that while Senate Democrats are more closely aligned than Republicans in their voting patterns so far in 2013 — Wilson notes that 22 Democrats have voted exactly the same on every vote this year — there are very few ties between the two parties. The two members sitting in the middle are Republicans Sens. Lisa Murkowski (Alaska) and Susan Collins (Me.), two of the noted moderates in the chamber.
Then there is the strange case of Louisiana Sen. David Vitter (R) and New Jersey Sen. Frank Lautenberg (D). Neither man has voted with any other senator more than 75 percent of the time during 2013. (Lautenberg, who is retiring in 2014, hasn’t even voted with any other Senator 65 percent of the time.)
The most obvious storyline from the amazing tool Wilson has built is that the two parties in the Senate have, at least by their voting records in 2013, almost nothing in common. That affirms the widening partisan divide that we’ve observed in the Senate and in politics more broadly over the past few years.
Fiddle around with Wilson’s infographic. It’s a great tool that can spawn a thousand insights into how the Senate works (and doesn’t). What’s yours?
The Fix