On Psychometrics (In Progress)

The following is still being written. As such, take it as an amalgam of ideas rather than as a coherent overview.

Everything you are can be summarized using five, sixteen, or potentialy hundreds of factors. These factors are called psychometrics and they are used to quantify the differences between us. While we currently employ this device across education, employment, and the treatment of mental health—as a non-exhaustive overview—the evolution preceding our modern method was fraught with purposeful segregation and inefficacious methods. To do this timeline justice, and adeptly illustrate how our minds went from mystery to mathematics, you must first comprehend pre-contagion medicine.

Where many interpretations of Greek antiquity cast their lens upon its proto-democratic structure or the foundational philosophical schools, perhaps the most interesting insight was the Hippocractic conception of temperament (NIH, 2013). This plainly stated that health and emotion are a product of four bodily fluids called 'humours'. When these humours become imbalanced, it was said to cause pathology within the body and mind. In response, the physician's role was to identify the symptom, say over-excitement or mania (sanguinity), then adjust the associated fluid—which in this case was blood. From this model, four core personality types were derived: choleric, melancholic, phlegmatic, and sanguine (as mentioned). For the first time in recorded history, humans could systematically differentiate between each other. And given the wisdom of hindsight, we now know the conclusion: humoural theory, which stood as an avatar for metrics of difference, initiated within society an intolerance for the exact idiosyncracy that itself instantiated.

When contagion (or microbial) based medical models were adopted, humoural theory was set to the torch. What remained in pre-psychological thought was a burning recognition of individual difference paired with a desire to explain its origin. Aptly named, next came 'On The Origin of Species', an enlightening jewel refined during the peak of Victorian colonialism (or precisely 1859) by the now household naturalist Charles R. Darwin. Here he famously proposed evolution, supposing that inherited difference cast across time would cause an increase in those traits which lent themselves to greater reproductive success. And since those traits are the expressed interaction between a changing environment and a functionally limitless gene pool, this model [evolution] permitted for all trait permutations across any scales of life. From here, joined by their aristocratic cohort, Darwin's polymathematic cousin, Sir Francis Galton, engaged this model to further understand human difference.

 Fig 1. Darwin's finches, as depicted in 'On the Origin of Species' (1859).

Fig 1. Darwin's finches, as depicted in 'On the Origin of Species' (1859).

Within his book Hereditary Genius (1869), which was also a precipitant for the later Eugenics movement, Galton investigated 'eminency' across and within lineages, quantified through lifetime achievement. Since, according to evolution, traits are genetically transmitted between relatives in proportion to their reproductive fitness, he [Galton] presumed that achievement would follow a similiar pattern. By analyzing 'eminent' British families (composed of artists, lawyers, scientists, etc.) he found the probability that each descendent might themselves be 'eminent'. The results were conclusive, yet challenged by confound, where as relatedness to another 'eminent' individual rose (on a pre-Hamiltonian metric), the probability of one's own eminency followed. This was reported as the hereditary transfer of a proto-intelligence, incorrectly measured using body length and reaction time, which seemed to determine life success (Goldstein, 2012). Or in simpler terms, Galton concluded that humans vary in predictable ways.

While Galton had ignored the impact of ascribed advantages over life, and consequently initiated a century long nature versus nurture debate, he had correctly conceptualized what would be called intelligence. Where in doing so, the shackles that bound his progress were uncovered, and upon inspection, appeared sealed by his reliance on family analysis. This was in part a self-imposed lean towards utilizing his eminent network, as per his colleagues claims, but fundamentally, his falter was an extension of the academy's statistical impediments. Hence when he [Galton] discovered the LaPlace-Gauss (normal) distribution through a translated Belgian text, read as Essays on Social Physics from Adolphe Quetelet, his statistical shackles crumbled, his fellows rejoiced, and the people of that tomorrow mourned. Using this distribution to interpret intelligence, he concluded, for each person with above average mental faculties there was (and still is) one that is equally below average. These less than mentally endowed individuals live in society, and given that they reproduce at a greater rate than the 'eminent' upper class, then society must be on course towards intellectual feeble-mindedness. In avoiding his ill-informed reductio ad absurdum, Galton devised a Machiavellian ultimatum: either to incentive propogation within the intelligent upper class or to prevent the opposite through force and legislation (positive and negative Eugenics, respectively). And despite these antics having directly sired the hate-filled Eugenics movement, Galton still founded an intellectual framework that made later humanitarian efforts possible. Of which was an innovative genre for evaluation that bubbled into Parisian pedagogy.

Having followed Galton's work on intelligence, one of Paris' leading psychologists, Alfred Binet, was himself sure that a general human capacity existed. Hence why when he [Binet] recevied an executive request—from the Public Instruction Minister of Paris—asking him to develop a standardized child intelligence metric, capable of locationg those with special needs, he believed the goal feasible. And as a product of the prevailing literature, of which dismissed Galton's reaction time-intelligence correlates and his emphasis on heritablility, Binet produced a novel test frame informed by a classic assumption: those who think clearer, faster, and retain for longer must possess a greater capacity (Minton, 1998). Along with his research partner, Theodore Simon, Binet set out to test precisely this. In doing so, they designed a 30-question index, which measured abstract and executive functions, that scaled in difficulty as the paired numbers rose. This scale was pivotal for evaluation, since as a child's age increased their relative aptitude for more difficult test questions indicated a "mental age". Should a child complete more questions than 75% of their cohort, then that child might be above average; inversely, should one complete less than, and too, proportional to the degree of incompletion, then that child's mental age might be below their true age. According to Binet, this disconnect in functioning was not only correctable and chronologically fluid, but also unattributable to either experiental acquisiton or "congenital idiocy"—two statements that American functionalists would come to disregard.


TO BE COMPLETED
American functionalist science was the nurturing domain psychometrics required to grow. Its western debut came from the Stanford–Binet test, a raw measure of intellect co-opted and modified by Lewis Terman, during the early 20th century. This term “intellect” was quantified as an intelligence quotient (IQ), or the g factor: for Charles Spearman’s general intelligence. With this metric, psychometricians could not only evaluate mental functioning, but also predict employment success, accession from poverty, law abidingness, and marital duration (Gottfredson). Much like Galton mentioned half a century prior, intelligence seems to quantify life success.


Just as life success amalgamates from a series of small victories, war success stems from triumph in successive battles. To win each battle, throughout World War I (WWI), the U.S. military had to select the most capable men. Their selection was done, by Robert Yerkes and associates, using two altered Stanford–Binet tests—for literate and illiterate personnel (Carson, 1993). Each soldier was evaluated through three criterion: general knowledge and logic, literacy (depending on their testing group), and numeracy. With these data, an ordinal ranking of soldiers was formed. Those who scored high became officers, those with average scores made suitable privates, and the bottom fraction were discharged for ineptitude. This was the start of standardized testing. For a test to truly standardize, the distribution of scores must cluster around the mean. The original Stanford–Binet test was criticized for not accomplishing this: they identified the average soldier as being of “below average intelligence” (Buchanan, Finch, 2005). By the very definition of average, any metric with significant incongruency between their average score and mean score is not valid. For this reason, future psychometricians placed the normalization of data central to their practice.


The Weschler Adult Intelligence Scale (WAIS), formed by David Weschler just prior to World War II, best demonstrated how psychometrics should be normalized. This was through the creation of an intelligence point-scale that, when administered, produced data remarkably similar to a normal distribution. With this came the power to not just infer the relative standing of a datum within the population, but to also make conclusions about unknown samples. Every subsequently potent psychometric has been modelled after this golden standard. Featuring the same methods which were constructed to evaluate intelligence, notably Spearman’s factor analysis (used to consolidate the g factor) and the normalization of data, the study of personality was fashioned to quantify human thoughts and behavioural patterns. The primary assumption of personality metrics, called the lexical hypothesis, was that our language evolved specific words because they described aspects of ourselves. If we factor out the co-occurrence and meaning between these words, then the result is five dimensions which describe every possible human characteristic: (i) extraversion, (ii) agreeableness, (iii) conscientiousness, (iv) neuroticism, and (v) openness.


The “Big Five” is a summative representation of where psychometrics currently stand: as potently predictive measure of social and life success, only trailing behind intelligence. This ability, as showcased in WWI, can and does effectively select between personnel for corporate, educational, and mental health needs. But these are mere surface level applications. Psychometrics, both in history and present application, are infinitely more complicated than outlined. They are the product of centuries of brilliance which has been funnelled into quantifying the metaphysical nature of humanity. This foray should not be taken to definitively describe the development and impact of psychometrics; instead, to turn your eyes towards a field worth viewing.

CourseBrent Urbanski