Tuesday, March 20, 2018

Research and Publishing. 1: Why Research, Why Publish?

Psychology is a very young science and there is much still to be learnt, in fact most of psychology is not well understood yet. And progress towards better understanding is made through research. This is not simply an academic affair, better understanding of psychology will ultimately improve lives, as trauma, depression etc. become better understood, and better treated.

By research I mean collection of new information, this is not simply the research that one might do for an essay, searching Google, but collecting new information from the real world with surveys, experiments etc. This is why research skills are so important to psychologists. To be a good psychologist one must, at the very least, be able to understand research. Ideally, good psychologists would also be skilled at doing it. Without such research skills it is impossible to be truly evidence based, and if psychologists are not evidence based then they are at serious risks of falling into quackery and pseudoscience. This, unfortunately, characterizes much of current Ecuadorian psychology.

It is therefore imperative for Ecuadorian psychologists to become research active. This is not easy, as many undergraduate psychology degrees in Ecuador lack the capacity to offer strong training in research skills, and there are few postgraduate options. This is unfortunate, as in other counties, such as the UK, psychology graduates are usually the most research-skilled graduates of all. But the situation here in Ecuador will not improve unless we push for it. And the fact is that there are some psychologists here with strong research skills. At the Quito Brain and Behavior Lab we have lots of collaborations with professors both within and without Universidad San Francisco de Quito. We also have lots of research assistants and interns, mainly doing unpaid work. But by collaborating and assisting they are all learning about research and developing skills. The message is that you don’t necessarily need formal training to get into research.

Research when it is performed, must then be published. Research that is not published to a global audience is not of much use. And publication should be in journal articles. I’ll discuss the issues of where, e.g. which journals, in the next blog post. But the for the moment it is important to understand that journal articles are the basic method of transmitting high-level research. Research published  by the London School of Economics suggested that in their survey of social scientists, about 63% or all publications are journal articles, and only about 17% are books. The other 20% is made of various other outputs such as ‘Working and Discussion’ papers. For scientists, including psychologists, I suspect the percentage for journal articles would be even higher. Psychology students should be reading journal articles, and professional psychologists should be writing journal articles.

Academic journals are the most important source of information for academics, including psychologists.

Journals are the basic source because they are peer-reviewed. That means they go through a tough process, being reviewed by anonymous experts from around the world. The majority of articles that are submitted to journals are rejected. And those that are accepted are usually only accepted after being revised based on the anonymous experts’ criticisms, sometimes with several rounds of revisions. It is the strict checking process that means journal articles are more reliable than any other sources of academic information. Newspapers and magazines are written for entertainment not information, and books in general are less trustworthy than many people realize. It is now recognized for example that many, many psychology textbooks contain grossly incorrect descriptions of basic psychological studies and phenomena.

Journals are therefore the basic vehicle of academic research. And researchers are judged mainly on their journal articles. I did my PhD at the Institute of Neurology in London. At the time I was there, the head of the Institute was Professor David Marsden. He was a remarkable scientist-practitioner. and it is said that from the date he graduated as a doctor to his death at age 60, he published 740 journal articles, 208 book chapters, 76 reviews and 100 research letters. An average of one publication every 12 days. This is a truly exceptional research output. Some academics never publish anything in their entire careers. It is notable too that David Marsden was also a clinician. He provides a fine example that clinical and academic are not polar. It may be that clinicians working solely in clinical practice don’t need to research, but those clinicians that also work at a University are also consequently academics. And academics should be involved in research. The point is that academics, whether clinical or not, are judged on their research output, mainly concerning their journal articles.

I have a book on Behavioral Neurology that was presented to David Marsden when winning the American Academy of Neurology Norman Geshwind Award. This is the inscription in the book that is also signed by many leading neuroscientists. An inspirational scientist-practitioner, David Marsden died suddenly just six months after receiving this award. 

So how do we judge research output? People used to talk about impact factors of journals, as a proxy measure for the quality of publications. But really the academic world has moved on from impact factors anyway. Nowadays individual journal articles are all digitally linked together on the internet via their reference lists. It’s easy to see how many times an article has been cited by other journal articles or books. Several different databases calculate this data and display it publicly. The most obvious example is Google Scholar. So rather than look at the impact factor of a journal where a piece of research was published, we can actually see how useful the individual article in question has been. And there is a lot of variation. Some journal articles never, ever, get cited by anybody. Some fly and are cited hundreds of times each year.

Google Scholar shows how many cites every article or book has. We can see in this example that the article by K. Anders Ericsson and Herb Simon was very successful, having been cited 5,688 times since it was published in 1980, which is more useful than knowing the impact factor of the journal. 
So, the real measure of an article’s success, and hence the author’s success, is how many times it has been cited. This is much better than simply looking at the number of articles published, or the impact factors of the journals, or the related metric of whether it’s Q1, Q2 etc. Now we can get a decent estimate of the quality of any individual academic’s research output.

We can look at how many times an academic has been cited. That is a good indication of their success in research. An interesting and commonly used metric is the h-index. This is a single number that captures both the number of articles somebody has published, and how well cited they have been. It is calculated as the highest number of articles that have been cited that same number of times. For example, at the time of writing, I have about 45 published journal articles. Some are highly cited and some are not, overall, I’ve been cited 1,661 times, but I have 18 articles that have all been cited at least 18 times each. My h-index is therefore 18.

This h-index is now a common way to evaluate academics. I’ve been to international conferences in which when the next guest speaker is introduced, their h-index score is publicly announced. By this metric, the most success psychologist in the world has been Herb Simon, who, at the time of writing this, has been cited 323,706 times. He has an h-index of 172. Puts my h-index of 18 into perspective. The research described above from the London School of Economics suggested that in social sciences in general the h-indices of university professors are quite low, ranging from an average of about 2.2 for Law professors to 7.6 for Economics professors. No data was given for Psychology Professors but I’d guess that it may be higher, given the significant research culture in Psychology.

Herb Simon was a truly remarkable academic. A leader in psychology, economics and artificial intelligence, he received a Nobel Prize in 1978 and is the most cited psychologist ever based on his h-index. I attended  the Cognitive Science Society meeting in Edinburgh six months after his death.  Several delegates there were in tears when talking about him. 
So why research and publish? Because it makes psychology better and makes us better psychologists. And it's what academics do. It's that simple.

In the next blog post I'll deal with the issue of where to publish.

Wednesday, March 14, 2018

'Street Children' of Latin America


One of my research interests is on how socioeconomic factors interact with cognitive and brain development. I’ve done studies with homeless adults in the UK and Japan, and have a study planned here in Ecuador. However, in Quito, the most conspicuous aspect of socioeconomic deprivation from a psychological perspective is the ‘street children’. This is in fact a common feature of several Latin American cities.

A street child that I photographed in AsunciĆ³n, the capital city of Paraguay. Children selling things to motorists is a common feature of life in many Latin American countries

Street children are a very poorly defined group, and perhaps the term shouldn’t be used at all. It really just means very poor young people who spend a lot of time unsupervised in urban environments. They are not necessarily homeless, and in fact most may have homes to go to and may even be attending school. In Quito the reason for being in the streets is usually to earn money, selling candies on buses and in bars, or shining shoes. I published some musings on this issue of definition a couple of years ago on Favelas@LSE (Pluck, 2015a).

Shoe-shine boys in Quito's Centro Historico. Many 'street children' are really working children. 

Nevertheless, the term sticks, as ‘street children’ is useful shorthand. The research I did was with students of Universidad San Francisco de Quito who helped as research assistants. We interviewed 37 former street children attending an educational program in south Quito. The first part of this was an evaluation of traumatic experiences. That revealed a very high rate of exposure to violence, and consequent post-traumatic stress disorder (PTSD). In fact, about 60% of the children we spoke to met criteria for PTSD (Pluck, Banda-Cruz, Andrade-Guimaraes, Ricaurte-Diaz, & Borja-Alvarez, 2015).  For that part of the research I was interviewed by Universidad San Francisco de Quito and they put a video clip about the research on YouTube.

A video about my street children research made by Universidad San Francisco de Quito

Next we collected data on a group of 26 ‘control’ children who had never been street connected. Both the street children and control children were assessed with several tests of cognitive function. These included a test of fluid intelligence, a test of visuospatial ability, and two tests of executive functioning, the Towers Test and Design Fluency. The reason that we were interested in these functions was that some of the anthropological literature has suggested that street children may fair remarkable well, and may even develop in some ways better than children who stay at home (in probable poverty and possible abuse). On the other hand, the medical literature on substance abuse, psychological trauma, exposure to violence etc. suggests that being street connected is particularly toxic to child development. A literature review that I did a couple years earlier tended to agree with that, in several studies from around the world, samples of street children were shown to perform poorly compared to non-street-connected children (Pluck, 2013).

Another photo of a child street vendor in north Quito, this was featured on the front cover of the Psychologist magazine. Inside was a review article titled 'The Street Children of Latin America' (Pluck, 2015b)

Our own data from Quito tended to agree with the medical view, the street children that we encountered scored quite badly on all of the cognitive tests, significantly worse than the control group. However, we did find some evidence that executive functions might be relatively preserved (Pluck, Banda-Cruz, Andrade-Guimaraes, & Trueba, 2018). This partly agrees with the anthropological perceptive, that the special challenges of being in the street environment as a child may drive the development of some abilities.

We also collected data on Theory of Mind ability in the same sample of street children. Those tests were measures of mentalizing, i.e. the ability to understand the mental contents of other people. We hypothesized that this could be enhanced in children who have to live in a competitive adult world. However, that data has proven quite a challenge to analyse. We are now collecting extra control data and hopefully should be able to add something new to the research literature soon.

Pluck, G. (2013). Cognitive abilities of ‘street children’: A systematic review. Chuo Journal of Policy Sciences and Cultural Studies, 21, 121-133. PDF


Pluck, G., Banda-Cruz, D. R., Andrade-Guimaraes, M. V., Ricaurte-Diaz, S., & Borja-Alvarez, T. (2015). Post-traumatic stress disorder andintellectual function of socioeconomically deprived ‘street children’ in Quito,Ecuador. International Journal of Mental Health and Addiction, 13(2), 215-224. PDF


Pluck, G. (2015b). The 'street children' of Latin America. The Psychologist, 28(1), 20-23. PDF

Pluck, G., Banda-Cruz, D. R., Andrade-Guimaraes, M. V., & Trueba, A. F. (2018). Socioeconomicdeprivation and the development of neuropsychological functions: A study with“street children” in Ecuador. Child Neuropsychology, 24(4), 510-523. PDF