Keith Schnakenberg and I continue work on a paper in which we develop a measure the unobservable level of respect for human rights. I blogged about it earlier here. In the new version of the paper we build upon existing insights in the Bayesian measurement literature to develop a dynamic ordinal item-response (DO-IRT) model. We then assess the validity of the estimates obtained from the latent variable generated from the DO-IRT model with the estimates from an ordinal item-response (O-IRT) model and the original additive human rights scales. Below is a plot of the estimates of the latent physical integrity variable (see the paper for a similar plot of latent empowerment variable):

DO-IRT physical integrity latent variables estimates for 192 states in the year 2008. Blue dots are point estimates (posterior means) and orange lines are 95% credible intervals.

Below is the current version of the abstract:

We develop a Bayesian statistical technique, the dynamic ordinal item response theory model, in order to estimate latent human rights variables from the two ordinal Cingranelli and Richards (CIRI) human rights indices (physical integrity rights and empowerment rights).   The dynamic ordinal item-response model builds upon the ordinal item response theory model and dyanimic item response model.  All of these models explicitly account for the measurement error that results when different component indicators are aggregated together.   We present both models, which each produce point estimates and credible intervals for latent physical integrity respect and latent empowerment respect.  We assess the validity of these latent variables in relation to each other and with existing human rights scales.  We then present a simple method that allows measurement error to be included in the uncertainty of causal estimates and also demonstrate the utility of the new latent human rights variables by replicating a recent analysis that uses ordinal CIRI human rights variable as two of the main explanatory variables of interest.  We conclude by recommending the inclusion of the new human rights scales in studies that posit a causal effect of human rights abuse on some outcome of interest. 

# —– naSortMatrix function —————————— #
# —– by Chris Fariss ———————————— #
#
# Takes a data frame or matrix as an argument
# and returns another matrix after moving any NAs
# within each matrix row to the right of the data
# in that row without changing the order of the
# data. Also calculates the time taken to complete
# the sort and saves it as a global variable:
# naSortMatrix.time
#
# Just copy and paste this during an R session to
# use the function
#
# ——————————————————– #

naSortMatrix <- function(x){
# time at start of function
d1 <- Sys.time()

# convert data.frame to matrix
data <- as.matrix(x)

# transpose matrix and then turn it into a vector
data.vector <- c(t(data))

# create a matrix of observation IDs and then transponse and turn it into a vector also
index <- c(t(matrix(c(1:nrow(data)), nrow=nrow(data), ncol=ncol(data), byrow=F)))

# combine the index vector and data vector into a long matrix
data.long <- cbind(index, data.vector)

# sort long matrix by index and the data vector. this will place all of the NAs at
# the end of each panel in the long matrix and keep the IDs from changing position
newdata <- data.long[order(data.long[,1], data.long[,2]),]

# transform the long matrix back into a wide matrix with the same observation order
out <- matrix(newdata[,2], nrow=nrow(data), byrow=T)

# calculate time at end and the difference between time at start as a global variable
d2 <- Sys.time()
naSortMatrix.time <<- d2-d1

# return the sorted matrix
return(out)
}

# —– end file ————————————— #

On Tuesday at the HNG meeting I’ll be discussing three experiments I’ve conducted on the acquisition of social network structure. For a preview, you can read my submission to CogSci 2011 which discusses one of the experiments.

Using a corpus of digitalized texts of The Google Books Project (i.e. the offline version of Google Trends), Michel et al. (2011) propose a new approach for quantitative investigation of culture. Their method may have broad impacts on various disciplines such as “lexicography, the evolution of grammar, collective memory, the adoption of technology, the pursuit of fame, censorship, and historical epidemiology.”

We should do one of these for the picture friend network!  Read about the image here.

Traditional game theory assumes that the level of recursive belief inference is infinite when people choose their strategy as a result of guessing the others’ strategies. For example, in Keynesian beauty contest where all participants are asked to pick a number between 0 and 100 and win if one is the closest to 2/3 of population average, Nash equilibrium predicts all players should chose 0 since recursive inference about others’ preference will decrease the value of your choice, and eventually reach the minimum possible value. However studies in behavioral economics have found that the degree of recursion is bounded to smaller values.

By running a 2 dimensional stag-hunt game, recent fMRI experimental study done by Yoshida et al. demonstrates that people vary their level of inference depending on their partner’s past strategic profiles. Imaging result shows that prefrontal cortex region is subdivided by its roles for encoding uncertainty of inference of partner’s strategy and inferring the degree of recursive inference.

McKenzie is a professor in the Department of Psychology and is also on faculty at the Rady School of Management.

Check out his research.  It’s pretty interesting.

http://psy2.ucsd.edu/~mckenzie/

http://management.ucsd.edu/faculty/directory/mckenzie/

Nicholas Christakis and I tackle this question with a little from a little help from Alyssa Milano….

http://www.cnn.com/2010/OPINION/11/16/fowler.christakis.tweeting/

Evolution of Coordination in Social Networks: A Numerical Study

Coordination games are important to explain efficient and desirable social behavior. Here we study these games by extensive numerical simulation on networked social structures using an evolutionary approach. We show that local network effects may promote selection of efficient equilibria in both pure and general coordination games and may explain social polarization. These results are put into perspective with respect to known theoretical results. The main insight we obtain is that clustering, and especially community structure in social networks has a positive role in promoting socially efficient outcomes.

More below:

Time scales of epidemic spread and risk perception on adaptive network

Incorporating dynamic contact networks and delayed awareness into a contagion model with memory, we study the spreading patterns of infectious diseases in connected populations. It is found that the spread of an infectious disease is not only related to the past exposures of an individual to the infected but also to the time scales of risk perception reflected in the social network adaptation. The epidemic threshold $p_{c}$ is found to decrease with the rise of the time scale parameter s and the memory length T, they satisfy the equation $p_{c} =\frac{1}{T}+ \frac{\omega T}{a^s(1-e^{-\omega T^2/a^s})}$. Both the lifetime of the epidemic and the topological property of the evolved network are considered. The standard deviation $\sigma_{d}$ of the degree distribution increases with the rise of the absorbing time $t_{c}$, a power-law relation $\sigma_{d}=mt_{c}^\gamma$ is found.

Opinion formation and cyclic dominance in adaptive networks

The Rock-Paper-Scissors(RPS) game is a paradigmatic model for cyclic dominance in biological systems. Here we consider this game in the social context of competition between opinions in a networked society. In our model, every agent has an opinion which is drawn from the three choices: rock, paper or scissors. In every timestep a link is selected randomly and the game is played between the nodes connected by the link. The loser either adopts the opinion of the winner or rewires the link. These rules define an adaptive network on which the agent’s opinions coevolve with the network topology of social contacts. We show analytically and numerically that nonequilibrium phase transitions occur as a function of the rewiring strength. The transitions separate four distinct phases which differ in the observed dynamics of opinions and topology. In particular, there is one phase where the population settles to an arbitrary consensus opinion. We present a detailed analysis of the corresponding transitions revealing an apparently paradoxial behavior. The system approaches consensus states where they are unstable, whereas other dynamics prevail when the consensus states are stable.

Understanding why people vote is a central concern of political science and democracy. Since elections are the central institution of a democratic society and since voting is the most widespread way of participating in an election, this centrality is quite sensible, obvious, and well-deserved. Despite this, we still don’t know a lot about why people vote.

I wanted to flag two new publications that go a little ways to explaining this better. The first is a paper I’ve just had come out in the CJPS. It’s largely derivative of Fowler’s early work (some of it with Kam) on the relationship between altruism and voter turnout. The second is a joint work with Fowler and Dawes (who else?!?).

Some researchers at the University of Indiana have created a website that lets users track politics-related tweets in real time. The site includes trends over time, network visualizations, among other things. This is a great example of the kinds of things one could do research on using the (freely available) data from Twitter’s API. While I don’t have the programming skills to do anything useful with the API, I can get some data out of the API using the R package for Twitter.

A recent NY Times Article reminds us of the effectiveness of social pressure on political behavior.  Check it out.

If you wondered like I did what “The Algorithm” was for Facemash in the movie The Social Network, wonder no more!  It turns out to be nothing more (or less) than the Elo rating system.

I blogged a little about it here.  The math is even there if you want to try it out.

Also interesting is the Elo vs. the Rest of the World competition going right now at Kaggle.  Also also interesting is the fact facemash.com is for sale.

Malcolm Gladwell has a new piece in The New Yorker on social media and activism. I have linked to it here.

In the article Gladwell makes a lot of assertions about how people use social media and what it is/isn’t useful for doing. Most of these assertions are not based on research; rather, they seem to be based on what he assumes about how Facebook and Twitter are used. This is a great article to get some hypotheses about how people actually use these types of media!

Here’s a heads up about a new database, just in time for the campaign season to kick into full swing: the Government and Social Media Wiki. It tracks which government officials and offices use social media sites like Facebook, Twitter, YouTube, and Flickr. The database includes members of the House and Senate, some congressional committees, federal agencies, governors, and even some candidates for House and Senate. So if your research requires you to keep tabs on Colorado Senate candidate Ken Buck on Foursquare or stay up to date with Senator John Thune’s MySpace page, here’s an easy way to do it.

The current issue of Cognitive Neuroscience is a special issue on the neuroscience of consciousness. Of particular interest is the article by Victor A.F. Lamme, “How neuroscience will change our view consciousness“. From the abstract:

…the study of consciousness is dominated by what we know from introspection and behavior. This has fooled us into thinking that we know what we are conscious of. …in fact we don’t know what we are conscious of. …The exercise is an example of how neuroscience will move us away from psychological intuitions about consciousness, and hence depict a notion of consciousness that may go against our deepest conviction: “My consciousness is mine, and mine alone.” It’s not.

The rest of the issue is behind a paywall (UCSD has institutional access), but this article freely available to everyone.

Political scientists don’t typically have to worry about the ethical implications of their research the way a nuclear physicist might. After all it’s hard to see how an improved understanding of the differences between presidential and parliamentary systems can turned into a weapon.

The turn towards brain imaging, genetics and social networks does however raise a troubling issue with how our work might be used to diminish the privacy and independence of people in society. A perfect example of this can be found right here in San Diego with No Lie MRI, which uses fMRI to offer “truth-verification” services for $5k a pop.

Fortunately, the use of such technology is inadmissable in court, thanks to the Daubert Standard; however, this relies on the fact that such use of fMRI has heretofore been limited to laboratory conditions, and hasn’t been well-tested “in the field”. It’s not clear how long this limitation will remain.

Something for all of us to think about in our own research efforts.

Population dynamics of human emotion has clear cycles!

http://www.ccs.neu.edu/home/amislove/twittermood/

Workshop on Social Media Analytics will be held on July 25, 2010 in Washington, DC in conjunction with The 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD 2010).

http://www.kdd.org/kdd2010/

What we do when we confront multiple contradicting ideas? The theory of cognitive dissonance asserts that people tend to manipulate their preference, attitude or opinion to evade such uncomfortable situations. Psychologists developed a standard experimental setup to assess the presence of such tendency. In this setup, the experimenter asks subjects to rank N objects according to their preference and offers n<N objects to them. After the transaction was made the subjects are asked to re-rank N products. Many experimental studies have found that the subjects are likely to alter their preference ordering. That is, they tend to rank those n objects higher afterward. Two fMRI studies (Sharot, Martino and Dolan 2009; Veen et al. 2009) revealed that dorsal anterior cingulate cortex anterior insula, caudate nucleus and amygdala underly certain behavior.

In a recent Brevia published in Science, Lee and Schwarz report that ordinary hand washing task removes the consequential preference shift caused by cognitive dissonance. This study extends the findings of previous studies which examined the role of physical cleansing on compensatory behavior and moral judgement, and demonstrates that it also has implications on people’s preference consistency. The link found in these works may elucidate result of voters’ physical and physiological activities on their political attitudes and decisions (see Mullainathan and Washington on cognitive dissonance in voting).

http://www.templeton.org/reason/

I ran across this corpus of Congressional speech that may be useful to some in the group.  Here is a brief description:

This data includes speeches as individual documents, together with:

• automatically-derived labels for whether the speaker supported or opposed the legislation discussed in the debate the speech appears in, allowing for experiments with this kind of sentiment analysis
• indications of which “debate” each speech comes from, allowing for consideration of conversational structure
• indications of by-name references between speakers, and the scores that our agreement/disagreement classifier(s) automatically assigned to such references, allowing for experiments on agreement classification if one assigns “true” labels from the support/oppose labels assigned to the pair of speakers in question
• the edge weights and other information we derived to create the graphs we used for our experiments upon this data, facilitating implementation of alternative graph-based methods upon the graphs we constructed

The third bullet seems like it would be of particular interest.

In my data mining class we are not using this corpus, unfortunately.  But, if you want to know which words most likely indicate an unfavorable movie review, I should have a classifier that will tell you by next week.

An experimental study shows that even birds have hierarchical social orders among them. By using delayed correlation analysis on birds’ flight directions, the authors uncover a weighted directed influence network of the bird flock. The topological structure of the network validates the presence of distinct asymmetric hierarchy within the bird community. This result implies that the long-held proximal interaction topology assumption in group dynamics simulation research should be replaced by the implementation of highly nondemocratic network structures.

Follow the leader from Science News on Vimeo.

There is an article in the latest edition of Science on social learning. The authors show that social learning is a key to success in their model. The article is interesting by itself, but there are many ways that the work could be extended through examinations of social learning from network, biological, and/or neurological points of view. Here is the abstract:

Social learning (learning through observation or interaction with other individuals) is widespread in nature and is central to the remarkable success of humanity, yet it remains unclear why copying is profitable and how to copy most effectively. To address these questions, we organized a computer tournament in which entrants submitted strategies specifying how to use social learning and its asocial alternative (for example, trial-and-error learning) to acquire adaptive behavior in a complex environment. Most current theory predicts the emergence of mixed strategies that rely on some combination of the two types of learning. In the tournament, however, strategies that relied heavily on social learning were found to be remarkably successful, even when asocial information was no more costly than social information. Social learning proved advantageous because individuals frequently demonstrated the highest-payoff behavior in their repertoire, inadvertently filtering information for copiers. The winning strategy (discountmachine) relied nearly exclusively on social learning and weighted information according to the time since acquisition.