Human Sciences, Statistics, and R

January 6, 2013

The use of statistics has long been important in the human sciences. An early example is an analysis by William Sealy Gosset (alias “Student”) of biometric data obtained by Scotland Yard around 1900. The heights of 3,000 male criminals fit a bell curve almost perfectly:


Histogram © A. H. Dekker, produced using R software

Standard statistical methods allow the identification of correlations, which mark possible causal links:


XKCD teaches us that “Correlation doesn’t imply causation, but it does waggle its eyebrows suggestively and gesture furtively while mouthing ‘look over there.’”

Newer, more sophisticated statistical methods allow the exploration of time series and spatial data. For example, this project looks at the spatial distribution of West Nile virus (WNV) – which disease clusters are significant, and which are merely tragic coincidence:


Distribution of significant clusters of human WNV in the Chicago region, from Ruiz et al.

SPSS has been the mainstay of statistical analysis in the human sciences, but many newer techniques are better supported in the free R toolkit. For example, this paper discusses detecting significant clusters of diseases using R. The New York Times has commented on R’s growing popularity, and James Holland Jones points out that R is used by the majority of academic statisticians (and hence includes the newest developments in statistics), R has good help resources, and R makes really cool graphics.


A really cool graph in R, using the ggplot2 R package (from Jeromy Anglim’s Psychology and Statistics Blog)

An increasing quantity of human-science-related instructional material is available in R, including:

Through the igraph, sna, and other packages (and the statnet suite), R also provides easy-to-use facilities for social network analysis, a topic dear to my heart. For example, the following code defines the valued centrality measure proposed in this paper:

library("igraph")
valued.centrality <- function (g) {
  recip <- function (x) if (x == 0) 0 else 1/x
  f <- function (r) sum(sapply(r, recip)) / (length(r) - 1)
  apply (shortest.paths(g), MARGIN=1, f)
}

This definition has the advantage of allowing disconnected network components, so that we can use these centrality scores to add colour to a standard plot (using the igraph package within R):


Social network diagram, produced using R software, coloured using centrality scores

– Tony


Suicide and the Military

December 18, 2012

A recent (2011) report by Margaret Harrell and Nancy Berglass (“Losing the Battle: The Challenge of Military Suicide”) looks at suicide in the U.S. military.


U.S. military active duty suicide rates, compared to general U.S. population (from Harrell and Berglass)

The U.S. military has seen an increase in suicide among its serving and former personnel, with suicide now killing more troops than enemy fire does. Junior enlisted personnel appear to be most at risk. Among U.S. veterans, it is estimated that there is one suicide death every 80 minutes and, although only 1% of Americans have seen military service, veterans account for 20% of U.S. suicides. The U.S. Army in particular has seen suicides increase markedly since 2004.

From Vung Tau, riding Chinooks, to the dust at Nui Dat,
I’d been in and out of choppers now for months.
But we made our tents a home, VB and pinups on the lockers,
And an Asian orange sunset through the scrub.

And can you tell me, doctor, why I still can’t get to sleep?
And why the Channel Seven chopper chills me to my feet?
And what’s this rash that comes and goes,
Can you tell me what it means?
God help me, I was only nineteen.
– Redgum, I Was Only 19

Harrell and Berglass make a number of recommendations to help address this suicide problem, including that the U.S. Army establish a “unit cohesion period” on return from deployment. Stress factors among soldiers include encountering dead bodies – and this is a stress factor which may also apply to troops on humanitarian relief missions. A 2003 book notes that “there is growing evidence that the stress of peace support operations can be as psychologically damaging as conventional warfare.” Given these stresses, addressing the problem of military suicide will require recognising post-traumatic stress disorder (PTSD) as a “real injury.” Various helpful recommendations for dealing with PTSD have also been made in the past.

According to the Centers for Disease Control and Prevention (CDC) there are over 36,000 suicides in the U.S. each year (and almost half a million cases of self-inflicted injury). Risk factors include stressful life events and feeling alone – two factors common among post-deployment military personnel who have left their unit.

Australian Defence Force (ADF) personnel appear to think about suicide more than the general community. According to one study, 3.9% of the ADF had suicidal ideation. However, comparing the 8 suicide deaths per year of ADF personnel against the 2,300 suicide deaths per year in the general Australian community shows that ADF personnel are slightly less likely to die by suicide than their civilian counterparts. This may indicate that ADF suicide prevention strategies are having some positive effects.


Arlington National Cemetery, Virginia, USA

Some of the recommendations from Harrell and Berglass also appeared in a 2010 U.S. Military Task Force report. That report highlighted in particular the need for reducing the stigma of soldiers seeking help, and for removing cultural and organisational barriers to doing so. RAND has also produced a lengthy report on preventing suicide in the U.S. military. In its July 2012 special issue on the topic, Time Magazine printed some helpful advice for wives and husbands of military personnel – see here. The Australian Defence Force has a fact sheet here.

In the end, though, suicide is everyone’s business, for “no man is an island … every man is a piece of the continent, a part of the main … any man’s death diminishes me.”


U.S. Army Suicide Prevention Poster (2011)

This post is dedicated to all my friends who wear, or who have worn, a uniform, and to everyone who has been affected by suicide.

– Tony


Emergence, Intelligence, Networks, Agents: WEIN

December 1, 2012

Networks are ubiquitous, and often large. The WWW contains over a trillion pages. The Internet contains over 900 million hosts. Facebook has over 900 million active users. The human brain contains over 80 billion neurons. The world has over over 7 billion people.

The behaviour of such networks is determined not just by the behaviour of individual nodes, but also by the network topology. The famous six degrees of separation often hold, for example. Network dynamics and self-organization are also important.

From the nodes and links of these large networks, fascinating things emerge. Nobody expected what happened when a small internal network at CERN was scaled up to planetary size, for example. And somehow, the interactions of our billions of neurons make us intelligent.

The 5th International Workshop on Emergent Intelligence on Networked Agents (WEIN’13) will be exploring some of these phenomena at AAMAS in Saint Paul, Minnesota next May.

The WEIN 2013 Call for Papers is out, and submissions are due January 30. It is likely to be an interesting event, and one that should help to build bridges between the multi-agent system and complex network communities.


Bridging disciplines: the PLoS One Map of Science

– Tony


Calling the Race

November 15, 2012

The big winner from the US Presidential election has been Nate Silver and his 538 blog. On the eve of the election, Silver had calculated the following probabilities from polling data:

Silver also has a fascinating analysis of which polls were accurate, and which weren’t. In particular, polls based on calling landline telephones tended to seriously underestimate the Democrat vote. Voters without landlines are more likely to be young, urban, Black, Hispanic, strapped for cash, or some combination of the five, and all five categories are more likely to vote Democrat. There are lessons here for pollsters in other countries.

The following simple NetLogo simulation model (click on the image to run it) re-rolls Silver’s electoral dice, giving alternative outcomes – exactly the kind of simulation Silver actually did to support his 90% prediction of an Obama win:

Silver appears to take a Bayesian approach to statistics. A Bayesian has been described as “one who, vaguely expecting a horse, and catching a glimpse of a donkey, strongly believes he has seen a mule.” The legendary XKCD summarises the perspective quite elegantly:

– Tony


Benefiting from Board Games

November 5, 2012

Playing the Ticket to Ride board game

I’m a big fan of board games, especially the newer German-style board games, which are far superior to the games of my youth. This book argues the benefit of modern board games for learning and teaching, highlighting some of my favourites.

Board games are more than just entertainment. According to a press release, a lecturer at the University of Tennessee has won an award for using the Ticket to Ride board game to teach operations research to students. This makes good sense, since German-style board games tend to involve complex optimisation decisions.


Some Dominion cards (photo: Shannon Prickett)

Consider a massively simplified version of the enjoyable game of Dominion, for example. There are six kinds of card: copper money (costs $0, worth $1), silver money (costs $3, worth $2), gold money (costs $6, worth $3), estates (cost $2, worth 1 point), duchies (cost $5, worth 3 points), and provinces (cost $8, worth 6 points). The real game has many other interesting cards, but even this simple parody is non-trivial.

At each turn, the player draws a hand of five cards (from a deck of initially ten), purchases a new card, and discards the hand. When the deck is empty, the discard pile is shuffled to form a new deck, so each purchased card will be “used” multiple times (which is why it’s sensible to spend $6 on a gold card worth $3). However, only the green cards are worth points – as with many German-style games, the money does not directly contribute to winning the game. On the other hand, purchasing many green cards reduces the chance of a five-card hand containing much money.

One strategy is to only purchase money or the valuable province cards. A simple simulation of the game shows that, after 50 turns, this results in an average score of 112 points. In contrast, a strategy of preferring to buy green cards gives an average score of only 39. However, “switching” from one strategy to the other does best of all, with an average score of 128 when the “switch” is made at turn 35 (and an average score of at least 120 when the “switch” is made somewhere between turns 25 and 45). In other words, winning requires optimising when the strategic “switch” is made.


Average scores, as a function of when the strategic “switch” is made. Switching at turn 35 is best.

For full-blown German-style games, the optimisation problems are more difficult. As this podcast argues, they are often in the difficult class of problem called NP-complete. These are hard enough to challenge both a human and a computer.

As I’ve said before, I also have a long-standing interest in collaborative board games, such as Arkham Horror, Pandemic, and Lord of the Rings (for fans of Arkham Horror, here is one of my custom characters, with back story and marker). Collaborative board games offer an excellent way of both exploring and teaching teamwork, and a 2006 paper by José Zagal and others explores Lord of the Rings in this context.


Reiner Knizia’s Lord of the Rings collaborative board game

Good board games generate the level of engagement that makes wargaming work. When the team is attempting to solve a difficult optimisation or decision problem (as in SCUDHunt), things can get very interesting indeed.

– Tony


Driver Distraction – are we being distracted away from real solutions?

October 29, 2012
Person using cell phone while driving.

Person using cell phone while driving. (Photo credit: Wikipedia)

 

Driver distraction is a major cause of accidents on our roads.  More research into driver distraction is therefore welcome.

 

Traditional media available in vehicles, such as radios and entertainment systems, can affect visual attention, and the current use of GPS navigation assistance and personal communication devices have been repeatedly shown to interfere with the primary driving vigilance and motor control task.  Such research has informed road authorities to restrict their use, particularly with regard to mobile phone use and texting for example.

 

The introduction of Intelligent Transport Systems has the potential to overload the driver if such systems are not tailored to driver workload according to location, traffic density and other ambient conditions.  A more accurate and layered approach to driver workload and attention level can provide a structure upon which information can be conveyed appropriately and distractions reduced or minimised.

Hume Freeway looking south towards Victoria, r...

Hume Freeway looking south towards Victoria, running parallel to Albury railway station. (Photo credit: Wikipedia)

 

There are many instances however where use of communication devices in vehicles may actually help in vigilance (for example, long distance driving) and the ability to operate a mobile office whilst driving has undoubted productivity benefits.  The problem is that the driving environment can potentially change in milliseconds and cause an overload of the driver’s available perceptual and attentional resources.

 

The aspect of appropriate speed limits in long distance driving could also be re-addressed.  There has been a spate of single driver transport accidents on the Hume Freeway recently which have had tragic (and potentially disastrous consequences). Apart from driver fatigue and scheduling issues, is the posted speed limit too low to maintain sufficient driver arousal levels?  Long boring drives at inappropriately low speeds could also encourage further distraction (such as use of mobile devices) perhaps even increasing the potential danger.  This may not be ideal given that the severity of an incident at 100kph is still as catastrophic as one at 110 or 120kph.  If speed kills, then surely it follows that the only safe speed is zero. There is a balance between the efficient movement of goods and the consequences of error.  Does this mean that interstate transport needs to travel at 40kph so that any incidents that occur are relatively minor? Would this result in more incidents because drivers would be bored out of their minds whilst blowing out transport schedules? There needs to be an open discussion by professionals in the field as to assessment of all of the risks rather than the current thinking that slowing traffic is the only solution as seems to be the case at present.  The posted speed relative to the road design and condition could be reducing driver arousal and performance level below the desired optimal range.

 

60KM/H Speed limit sign in Australia.

60KM/H Speed limit sign in Australia. (Photo credit: Wikipedia)

The Human Machine Interface (HMI) remains central to safe and effective vehicle operation as the information it provides will allow the driver to retain effective control of the vehicle and help influence or even determine their behaviour. Information flow to the driver must be intuitive, readily understood, timely, and be responsive to driver attentional and distraction state.  Unfortunately, there are many who have very little understanding of this requirement. For example, presentation of the “bells and whistles” mentioned in this proposed level crossing warning system  may actually distract a driver at the worst possible time and cause more problems than the technology is trying to solve. Human scientists understand the many facets that determine the best way of presenting information so that it is perceived, recognised and acted upon in an optimal manner.  If the proponent of this system had engaged human factors expertise in the first place or heeded their advice he may not have made such ill-considered comments. Perhaps he may get some guidance as to the importance of listening to human factors professionals when he presents his data at the upcoming ITS conference in Vienna. It would be good to have some critical assessment by any human scientists attending this conference of any actual (rather than derived or contrived) interactions that occurred during the trials which were conducted of this system.

 

The area of human interaction with technology is very complex and simplistic approaches (such as further unrestrained visual or auditory “bells and whistles blaring”) will rarely be the best solution.  All the more reason to design and test proposals from a human science standpoint, heed the results and incorporate them into any proposed system.  Which underscores the importance of the many respected facilities that incorporate human science input as a keystone of their research in ITS applications, vehicle design and driver behaviour.  Hopefully their research findings are weighted appropriately (ie. seen as valid and reliable) by the governing authorities as compared to those of the “bell and whistle” variety.

 

 

 

 

 

 

 

 

 


Lollies or Poison

October 16, 2012

Human factors covers many facets of human behaviour and interaction with the natural and constructed environment.  The issue of how to ensure that these environments remain safe for the younger and older members of our society is a case in point.Post image for Product Confusability: Tide Pods

Recently, a dishwashing detergent was packaged so that even adults could easily mistake them for food.  Young children have ingested these dangerous chemicals as the packaging is very colourful and resemble candy or lollies.  The harm done to these children and the increasing toll on the health system would surely outweigh the financial benefit to the manufacturer of this product, who obviously either didn’t think or was blissfully unaware of the consequences.  Perhaps the unwanted publicity of the dangers of this product, especially if there is an impact on their bottom line, may encourage them to design their products with more care in future.

All too often, human behaviour is not taken into account in the design of products.  No amount of urging customers to “be careful” will eliminate the danger posed by badly designed or labelled goods.  It is better business to design them properly in the first place.

When I was involved in training service and police dogs, there were five principles which used to guide us.  Firstly, Knowledge was required – if you didn’t know what you were doing, it made it difficult for the canine to know what was required.  Secondly, Repetition showed the required behaviour.  Thirdly, Patience in attaining the goal was essential.  Praise and Correction were the final principles, the former being more desirable and effective.  It seems that in the case of the design of the products described above, the first fundamental principle was lacking, as is commonly the case in many other applications where the knowledge and expertise of the human scientist is not sought or ignored.

This translates to the whole range of human applications.  Human scientists can provide critical knowledge but this fact is not often understood and can have great impact, both in terms of time and budget when ill informed decisions are made, and especially when the younger and older members of society are involved.

As has been previously posted, the initial BMW iDrive was a product which was installed in vehicles that the older demographic was more likely to acquire, which compounded the problems of poor initial design.  The older drivers did not have the digital savvy or knowledge that their offspring or grandchildren had and therefore the whole concept was flawed from the outset as it did not take into account the basic metrics of the human operator.  Of the five principles that I listed above, Patience is in short supply when you are battling traffic!

Which leads to the point made by the authors of the pieces – consideration of the special needs of the younger and older demographic is central to good design and outcomes going forward.