Paper aeroplanes in space

This post was inspired by a conversation I had with my friend Nikki about unemployment. After discussing the painful process of job seeking, she composed a sonnet in honour of her desire to work in fast food.
O, trap of grease
How you both fill and fulfill me
Thine stench and texture
Both excite me and envelope me with desire

Of course it was all very amusing, but it got me to thinking about how vile the process really is and hence, this post was born. As an ode to the odious process of job seeking, I have compiled a list of the four worst things about job seeking and unemployment.

Sending a job application is like throwing a paper aeroplane into the void of space and hoping it will crash land on an inhabited planet. You know when you write one that the chances of anybody reading it are vanishingly small, yet you still must spend hours hatefully crafting your work experience into action oriented sentences liberally peppered with offensively bland key words. You hope your flimsy creation will navigate itself through the minefield that is the key word seeking software. This software is like a rapacious, mindless beast, happily rooting for the truffles that are your words. Of course, everyone else's resumes are likewise splattered with these meaningless pieces of jargon so the truffles are more like undergraduate degrees, too common to be interesting to, or even acknowledged by most HR professionals. Which leads to depressing fact the first; It has been estimated that 75% of applications will not be acknowledged in anyway.

Your application has been successfully launched into the empty vacuum of space

Now I will acknowledge that unlike the majority of the statistics I provide on this site, I do not have an academically rigourous source for this statistic. The Googling I did however, kept coming up with this figure and it tallies pretty well with my own experience, so I invite you to think about whether it tallies well with yours as well. Considering that with an electronic application system it takes exactly zero effort to acknowledge you application and send out a generic email when the position has been filled, I think there is no excuse for this egregious breach of basic human etiquette. I may not be your ideal candidate, but if I have spent 3 or 4 hours on an application for your company, the least you can do is set up an automatic notification system to let me know I am going to continue to live on baked beans for at least another week. Psychologically speaking, this whole process of sending a resume and getting no response is totally non-reinforcing. In human and animal learning, if a behaviour (say posting a Facebook status) is consistently paired with something you like (Facebook likes), this will increase the likelihood of your performing the action in the future. But if you perform a behaviour and receive nothing at all consistently, you stop doing it. Even worse, people hate being ignored. It is why ostracising is such an effective social punishment. So hearing nothing becomes a punishment. The behaviour of submitting a resume actually starts becoming associated with a punishment, making you even less likely to submit resumes in the future. That is why studies have consistently found that the longer you are unemployed, the fewer job applications you do. This total lack of basic human etiquette is actually damaging to psychological well being and motivation.

Which brings us neatly to the second depressing reality about job seeking and unemployment. Long term unemployment is extremely bad for your psychological and physical well being. There are numerous studies showing that long term unemployment is associated with increased incidence of major depressive disorder, anxiety, psychosomatic symptoms, low subjective well-being and poor self-esteem (http://www.sciencedirect.com/science/article/pii/S0001879109000037, http://www.academia.edu/1213745/Associations_between_unemployment_and_major_depressive_disorder_evidence_from_an_international_prospective_study_the_predict_cohort_). They are also six times more likely to commit suicide ((Bartley et al, 2005) and one author has estimated that the effect of being long term unemployed is equivalent to smoking ten packs of cigarettes a day (Ross 1995). Stop and let that one sink in for a while. Unsuccessful job seeking is actually toxic. Of course, the correlation runs the other way too, with the unemployed more likely to engage in unhealthy behaviours (Waddell and Burton 2006). As Mansel Aylward, director of the Centre for Psychosocial and Disability Research at Cardiff University so elegantly said "“Sickness and disability are among the main threats to a full and happy life; work incapacity has the most significant impact on individual, the family, economy and society."

I love you sweet desk job

Of course to get a job, you must go through the process of applying for a job, which brings us full circle and to the third depressing reality about job seeking an unemployment: Resumes are written in a totally revolting and inauthentic style. There aren't many things as depressing as pretending to be enthusiastic about things no sane human could ever have a legitimate interest in.  Could anyone ever write the following sentence “I am passionately dedicated to statistical analysis and generating actionable insights” without wanting to punch themselves in the nose? I wrote that gem myself and felt dirty afterwards. The language you have to use in crafting these masterpieces of doublespeak. Resumes and cover letters require a revolting stylistic mix of self-aggrandisement and slavish servility that most people rightly feel is artificial and ridiculous. There is nothing authentic about the way you express yourself in a resume, right down to the bizarre syntax you have to use to being every sentence with an action word. Here are a few choice extracts from my resume:

• Pioneered several bespoke statistical calculators for use by all analysts

• Created innovative analytic technique for analysing data for a major medical device company

• Successfully presented results to a variety of international clients

Reading this makes me hate myself. It makes me hate anyone who could hire me after reading it.

This collection of words may be single handedly responsible for preventing intelligent aliens contacting earth

The final depression reality about the job seeking process is probably the most depressing one on this page. The system of resumes and cover letters can be as effective in picking the correct candidate as picking at random. That’s correct; depending on how you conduct the process, you are as likely to pick a good employee by flipping a coin or throwing a dart at a bunch of resumes stuck to the wall, as using the cumbersome HR machine. This finding comes from a meta-analysis examining the efficacy of a number of inputs into the selection process (Schmidt, F. L. & Hunter, J. E . (1998). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin,  pp. 262-274. As an aside, a meta-analysis is a sophisticated statistical technique that uses studies as individual points of data instead of individual people as a standard experimental study would. They are an extremely powerful technique that can give a good estimate of the real size of an effect in the general population.

This study of studies found that out of 20 possible information sources, the information contained in your resume including reference checks, job experience, years of education and interests ranked 13th, 14th, 16th and 17th at predicting job performance. The only things worse were your interests, graphology (the defunct science of analysing your handwriting) and your age. The best predictors were work sample tests (performing a test run of the main tasks you will be required to perform) and your general intelligence. Structured employment interviews came in third, with unstructured coming in 9th. In an ideal scenario, employers are picking the 13th-17th best predictors of job performance to whittle down a list of potential candidates, and the 3rd (or 9th) best predictors to make the final decision. You could set up a psychic hot line recruitment business and perform as well as the standard process for employee selection. On second thoughts, forget that, I have a brilliant idea for a new business.

Mirror, mirror on the wall, who should be my PA?

Beware of magic numbers

Every so often social media erupts with a new, numeric meme. This magic number purports to summarise in one easy, often round integer, some universal truth. The classic example is the “You only use 10% of your brain, we have all this untapped potential” hokum that you often hear regurgitated. My most recent numeric meme experience was “you have 35,000 thoughts a day”.
My Facebook and Twitter feeds started trending wit this claim and because I am a scientist, a data nerd and a profound killjoy, I decided I would investigate this little snippet. There are a couple of problems with any claim like this one;

1. How is a thought operationalized? In sciences like psychology, there aren't universal definitions the way there are in other sciences. For example, if I made the claim that on average people are 9 ft tall, you could whip out your ruler, gather a random sample of people and get measuring. You and I would have the same definition of one foot (i.e. 30.48 cm) and you would easily be able to demonstrate that my number is bogus. Since we all have the same rulers a claim like this is easier to assess objectively.

A thought however, is not an standardised unit. There aren't even very good definitions of a thought, much less guides on how to determine one discrete thought from another.
Wikipedia gives us this helpfully circular definition:
“Thought can refer to the ideas or arrangements of ideas that result from thinking, the act of producing thoughts, or the process of producing thoughts. In spite of the fact that thought is a fundamental human activity familiar to everyone, there is no generally accepted agreement as to what thought is or how it is created.”
What Wikipedia has generally correct, is that there is no generally accepted agreement on what a thought is. Certainly in cognitive psychology we wouldn't use the term ‘thought’ as an operational term, because it is too vague and unspecified.
Additionally, thoughts aren’t really discrete units like centimetres or feet. Human minds work using a kind chaotic spread of activation. What this means is that the activation of one idea spreads to semantically linked thoughts and concepts. This is why you can start talking about work, remember you had a coffee at work, think that you like coffee a lot more than tea, and start talking about tea in the next breath. Consider that chain of thoughts for a minute. How many thoughts is that? Is that one big thought about coffee, your experience of it today and your rating of it relative to tea? Or is it three thoughts, one for each concept? What about the sub-threshold thoughts that helped you link all those ideas together?

2. What is the source of the claim? If I am going to go around repeating an empirical claim, I want to make sure that the claim actually has some evidence to back it up. So the next step in my over-analysing of this popular meme was to look for the source of the claim. A fairly rudimentary search on Google Scholar confirmed my suspicions. There was definitely no published empirical work asserting that we have 35,000 thoughts a day, or any other specific number either. The closest I could come to was a nutrition study that aimed to measure how many diet related thoughts we have a day. Even this study, with its relatively limited scope had great difficulty in defining what a single food related thought was, and they essentially just ended up asking participants to mark down in a diary every time they thought about food. This approach would not really work for measuring how many thoughts of any kind we have in a day.

Imagine me asking you to write down in a journal every time you had a thought. The very act of thinking "I better write down all my thoughts today" is a thought. Even if you sat basically catatonic for the whole day, if you remembered the task at all, you would sit there tallying the whole day. Apart from being immensely impractical that kind of a study wouldn't tell us very much about the act of thinking in a natural setting.

Having established that there was no published empirical source, I kept digging. At this point, serious alarm bells started ringing. Not only could I not find any empirical source, I couldn't find any primary source of any kind. It seemed like this meme had just sprung into life of its own accord, infecting social media like a virus. Finally, after a few hours of digging, I found THE source. It was a promotional info-graphic which made a number of empirical claims, amongst them our 35,000 thought cap. Here it is for your viewing pleasure (and a link for those who are truly curious).

If you look bottom, you will see, in tiny pink writing, a handful of references. At this point I got excited and thought that maybe I had misjudged the magic number. Maybe there was some science behind the hype. Or not.

The first reference was http://sourcesofinsight.com/10-ways-to-defeat-decision-fatigue/, a self-help website article on the ways to help defeat decision fatigue. It outlined a list of fairly sensible time saving techniques like making lists and setting time limits on tasks. All very practical but sadly lacking in magical numbers. Certainly nothing worthy of meme-ification.

The second reference was http://seekingalpha.com/instablog/698556-jason-matias/377101-decision-fatigue-what-investors-should-know-about-the-science-of-decision-making, a blog post on an investment website which featured some very interesting and totally unsubstantiated claims like this gem “Ninety-five percent of the choices we make every day are managed autonomously by the filters we have developed though our interaction with the world and society. The other five percent of the decisions we make are made by our active self: our ego.” This article was full of magic numbers and certainly some of them might be meme-ifiable but sadly, my magic number was nowhere to be seen.

The next reference was http://www.sciencedaily.com/articles/a/amygdala.htm This was a link to a page describing the anatomy and function of the amygdala (it is involved in the processing of emotions, amongst other things). This was certainly a very science heavy link, giving some neuroanatomical authority to the claim, but was unfortunately also totally unrelated to the magic number.

Just when I starting to lose all hope I found this http://powerofpositivity.net/?p=349. I saw the number 35,000. I got very excited. My data-philic heart started to pound. The reference was a blog post by basketball coach, trainer, author and husband, Cornell Thomas and the key line, the source of all this meme excitement of social media had been found.

“A survey I read said the average adult makes about 35,000 decisions per day.”

And that was it. No reference. No link. No nothing, just this one sentence. Now it might be true that Cornell Thomas read just such a survey. Unfortunately without the source, this claim has all the empirical strength of a Cosmopolitan personality test.

To make matters worse, the info-graphic itself was part of a press release promoting….you guessed it, a decision making app. (For those who are interested: https://www.easilydo.com/). Hours of my time, thousands of Twitter and Facebook mentions, and this empirical claim is reducible to a marketing ploy. 

There is a moral in all of this of course, other than the fact that I probably need to get a hobby. When you hear any kind of empirical claim, don’t take it at face value. Don’t assume that it is correct, that is has been researched, or that the speaker even knows whether it is true or not. This is especially important now as Australia rolls into election time and politicians start flinging around “facts” like these ones.

“Most asylum seekers arriving by boat are economic migrants and on some boats 100% of the asylum seekers are in fact economic migrants and not genuine refugees.” - Bob Carr June 2013 (One of many sources)

This despite the fact that government records demonstrate that more than 90% of asylum seekers arriving by boat are found to be genuine refugees (Asylum seeker facts). Five minutes of research, and I mean easy Google research, can disprove many claims like this, but most people will never even take the trouble to fact check for themselves. This willful ignorance only allows politicians and other people to get away with all kinds of unsubstantiated claims.

We have the capacity for so many thoughts a day (although not necessarily 35,000), let’s use some of them critically.

 P.S At the time of writing this a new thought meme is circulating. Apparently 70,000 is the new 35,000.

Welcome, featuring thoughts on categorisation, equality and blank slate activism

Welcome to Science: Good Bad and Bogus, a name which I freely admit I borrowed from a very interesting History and Philosophy of Science course I took in second year university (thanks Dr. Slezak). The central goal of this course was to solve the demarcation problem, which for the uninitiated, is the problem of defining science. How do we carve up all the investigative disciplines into Science and Non-Science? This, like many other fascinating philosophical discussions, is essentially a categorisation problem. As human beings we have an innate and generally very adaptive desire to categorise the world into groups of A's and B's. We run into trouble of course, when the world isn't as neatly demarcated as the categories we are attempting to fit it into. We are especially bad at the boundaries of categories. Consider the very emotional abortion debate, which centres on when something is "alive" or has "personhood". In other words, when does an entity move from the category "Biological Mass" into the category "Human Being"? Nobody in the pro-life camp believes that all biological masses deserve moral consideration and nobody in the pro-choice camp believes that all human beings should be killed at whim. The disagreement comes when we try to agree on the category boundaries. 

This introduction nicely provides the background for the main topic of this post, which would be a coincidence too amazing to be believed, if I hadn't planned it that way. In recent adventures on the internet (especially on the forum I am a member of) there has been lots of discussion around equality. Naturally this thorny topic provides a great deal of discussion that can essentially be reduced to discussions of categories of people. Which categories are valid for making judgments upon? Which categories are even real? Should we be discriminating on the basis of categories at all? Interestingly, in the context of categorisation, discrimination just means learning to differentiate between members of different categories. Generally, we learn to discriminate based on features that are most predictive of differences between the categories. That is, we pick the features that most reliably signal category membership. And we tend to be very good at it. It isn't very often that we latch on to a feature that is totally non-predictive and persevere in using it (we do sometimes pick a feature that is less predictive, but correlated with, another more important feature).

With that in mind, is biological sex a useful predictive category on which to make judgments? Can gender equality be achieved whilst acknowledging there may be differences between sexes*? 

The debate centres on the following points:

1) Is sex a real (biologically extant) category?

2) Is sex a useful (predictive) category?

3) Is sex distinct from gender? I won't really be addressing this point, in this particular post.

Arguments around gender equality have typically centred on the first two points. It is articulated as follows; sex isn't a biological reality at all and doesn't predict any reliable differences (behaviourally, psychologically or otherwise). There is no sex; we are all the same, Q.E.D. equality. Leaving aside the fact that I believe I can empirically demonstrate that the first two points are incorrect, deciding that we are equal because we are identical is a disturbing and dangerous assertion. It is essentially placing the cause of equality on the incorrect notion that we are all born tabula rasa (blank slates). We are all identical to begin with; therefore we are all equal in our potential. The problem with this argument is twofold:

1) It is flatly contradictory to the pro-equality mantra that diversity is a wonderful, enriching thing (which I wholeheartedly believe). We can't be diverse if we are all identical.

2) It places equality on very unsteady ground, because anytime anyone demonstrates that there are in fact differences between people, it is seemingly falsified. 

People aren't entitled to equal treatment because we are all carbon copies of each other. We differ in numerous beautiful ways (height, weight, sex, intelligence, neuroticism, extraversion...) and many of these differences are very appropriately used to determine our suitability for jobs, partners, and hobbies. People are entitled to equal treatment because at the most important level, that of humanity, we are the same. We are all members of this superordinate category and we have decided together that membership means a right to certain standards of life. We might one day even extend this category to include members of other species, but sharing this high level category doesn't obliterate the important within category diversity. The reality is that like with all of our other categories, the world is far messier and shaded that the categories contained within our minds. Evidence suggests we can train ourselves into forming fuzzy categories with semi-permeable boundaries and these sort of categories hold out the best hope for us understanding and thinking about thorny issues. 

Since categorisation is an inescapable facet of human cognition that is often extremely useful, it seems perverse to force ourselves to completely ignore them. Instead, we ought to acknowledge that in the absence of all other information, they provide a statistical 'best-guess' of what a person might be like. This can, and indeed should, be refined as new and more person specific information comes to light. It is only when categories are overgeneralised and impervious to disconfirmation that they become stereotypes.

For the final say on this topic, I hand over to Steven Pinker, whose ideas have been instrumental in helping me form my own

"As many people have pointed out, commitment to political equality is not an empirical claim that all people are clones."

                                - Steven Pinker
 

*Whenever I talk about differences, I mean at an aggregate level. There are for example reliable differences in the average height of men and women, with men being statistically significantly taller on average. This doesn't mean an individual woman can't be taller than an individual man. This will always hold true for any differences discussed.