From Obscurity to Clarity in Psychometric Testing

Selected works of Professor Peter Saville

Peter Saville

• Publisher: Routledge
• Year: 2016
• ISBN:
  9781317585091
• Edition: 1
• Topic: Psychology
• Subtopic: History & Theory in Psychology
• Index: Psychology

The field of psychometrics has a long and varied tradition across the social sciences. A range of academics have sought to understand human consciousness more fully by statistical testing our abilities, personalities, attitudes and beliefs. But perhaps the area where psychometric techniques have had most impact on contemporary society is in employee recruitment, where a range of psychometric tests have become common-place.

Professor Peter Saville is one of the pioneers of adapting psychometric testing to the field of occupational psychology and human resource management. In a career of nearly 40 years, his work has been adopted by hundreds of public and private organizations, assessing the suitability of prospective candidates through a range of questionnaires and tests. In this anthology of his work, including both keynote conference address and journal papers, Saville provides a masterly overview of the field of psychometrics, and the key issues and questions that it raises.

An ideal companion for any student or researcher of HRM, occupational or organizational psychology, or applied psychology in general, Peter Saville's selected works represent the thinking of one of the most influential psychologists of our time.

PART I The origins of psychometrics

1 What's it all about?

DOI: 10.4324/9781315742083-1

Introduction

This chapter is based on a keynote speech which I delivered at a conference in South Africa in 2008:

Saville, P. (2008). Psychometrics in the workplace. Presented at the 14th South African Psychology Congress, Johannesburg, South Africa, August.

Here I will provide a brief introduction and overview of a number of concepts which will be discussed in greater depth throughout several of the chapters in this book.

Psychometrics in the workplace

There are a wide variety of different assessment methods that can be used. Just some examples include:

• Reading fluency.
• Spelling accuracy.
• Grammatical understanding.
• Verbal reasoning.
• Numerical reasoning.
• Abstract reasoning.
• Logical reasoning.
• Spatial reasoning.
• Mechanical reasoning.
• Clerical/error-checking ability.
• Manual dexterity.
• Physical strength.
• Tests of sensory abilities (e.g. sight, hearing).
• Application forms.
• Biographical data.
• Interviews.
• References.
• In-tray exercises.
• Group activities.
• Projective techniques (e.g. inferring characteristics about a person from how they interpret certain situations, images or even abstract shapes).
• 360-degree performance assessment (i.e. where various stakeholders all offer their assessments of one particular individual).
• Values/preference surveys.
• Personality/behavioural questionnaires.
• Observation/job simulation.
• Assessment centres (where multiple different kinds of assessments are used in one overall assessment period).

These various kinds of assessments have been used in countless different situations and applications. A number of them are to this day used in the workplace and some are used in other contexts too – educational or clinical environments, for example. Certain kinds of assessments are straightforward and transparent, whereas others may be more complex and require a greater degree of understanding to use and interpret appropriately.

We know, for example, that the observation of people at work has been used to assess many individuals in a vast number of different situations. A neurosurgeon once told me that in the 1800s, surgeons often insisted that their apprentices be present early on for especially gruesome operations such as sawing off somebody’s leg. The rationale was that if the apprentice fainted or couldn’t bear to watch, they wouldn’t be hardy enough to train as a surgeon and so had to be dismissed early on.

Such ‘job samples’ have also taken other forms. I once heard that in Japan sales applicants to certain jobs would be required to climb to the top of a hill or other high place overlooking a large number of people. They would then be asked to scream the name of the product that they would be required to sell. The rationale was that if the individual was too scared to do this, they wouldn’t be able to perform successfully in the role and so would have to be rejected. Other graduate applicants in Japan were reputed to have been required to clean a toilet as part of their application process. The rationale here was that this would demonstrate how motivated and detail-conscious they were. This particular group of assessors also observed the graduates during mealtimes. If they ate quickly and properly, they were kept in the applicant pool, but if they were messy or slow, they were rejected. In this case, their personality characteristics were being inferred from their eating style. Lastly, we also know that ancient Chinese civil servants were assessed not only on their intellect, but also on their creativity. This was sometimes assessed by giving them the first line of a poem or rhyming couplet and then asking them to provide the most original possible following line.

Speaking more specifically about ‘psychometric’ assessments, there is one particular sub-class of assessments which is often used in workplace contexts. So when we speak of psychometric assessments, what are we actually talking about? In broad terms, psychometrics are standardised assessments of psychological attributes. It can be useful to categorise them at the highest level into ‘can do’ assessments of maximal performance and ‘will do’ assessments of typical performance. In practical terms, this usually represents the distinction between assessments where there are right and wrong answers (e.g. IQ or ability tests) and those where there aren’t explicit ‘correct’ responses (e.g. personality or preference questionnaires).

Personality or behavioural questionnaires are among the most widely used workplace assessments and I’d like to take this opportunity to provide a brief overview of the history of personality research. ¹

One of the earliest known attempts to define personality came around 3,000 years ago, when Buddhist writers outlined the five most important characteristics of elephant trainers (surprisingly close to the Big Five model of personality discussed later) as health, confidence, wisdom, diligence and sincerity.

We also know that descriptions of personality go back to the ancient Greeks. Plato postulated the need for society to have governors who were leaders and philosophers. He indicated that these individuals required specific qualities to succeed; they must be truthful and not at any time be cowardly.

The great ancient Greek physician Hippocrates first proposed the four temperamental characteristics ‘Melancholic’, ‘Phlegmatic’, ‘Choleric’ and ‘Sanguine’ which he called ‘humors’. Hippocrates believed that these humors were caused by changes in certain bodily fluids, namely: black bile, blood, yellow bile and phlegm.

Hippocrates was therefore one of the first to conceptualise a personality theory. He proposed that only the proper mixture of these four humors would produce the condition of health. Temperament theory suggested that though the proportions of the humors may vary considerably from person to person, people could in essence be reduced to four types of temperaments according to the predominance of a given humor.

Since there were four humors, it was proposed that there could be four kinds of healthy equilibrium, not one, and that individuals could be subdivided into four psychological groups named after the prevalent humor: the sanguine, buoyant type; the phlegmatic, sluggish type; the choleric, quick-tempered type; and the melancholic, dejected type. These humors were, in turn, linked to the four fundamental elements of the universe: air, water, fire and earth, respectively. Hippocrates believed that personality and illness were dependent upon these four humors. Any imbalance in these humors would result in a personality problem or an illness.

Research has also brought to light that the Vikings actually had a competency-based system for selecting their leaders; it was not necessarily done through the blood line. As historian Paul Mortimer ² has explained to me, if a king’s son could not reach the standard required by the elders, they were put to one side and a more worthy individual would be put through the selection process, even if not of royal ‘blue blood’. Such selection processes included measures of bravery (they had to show a reputation for exceptional achievements), job knowledge (they needed to have absorbed education from an expert), presence both on and off the battlefield (they had to show signs of wealth, eminence and were often taller than average), eloquence and persuasiveness (they needed to be good orators), a keen intellect, evidence that they had helped develop the young, being generous (they circulated their wealth and accolades), but the capacity to be cruel to anyone who challenged them.

They had a clear benefits and rewards system in the workplace. As Paul Mortimer continued: ‘A stingy, niggling king did not last long.’ Although the Vikings were often regarded as a rabble when fighting, in fact they were immensely skilled and adroit at strategy, which made them a formidable fighting force. Lastly, I should add that it is also a myth that they wore helmets with horns – they did not!

We also know that people have tried to find links between the biological and the mental. Shakespeare (1599) wrote in Julius Caesar:

Let me have men about me that are fat;

Sleek-headed men and such as sleep o’nights;

Yond Cassius has a lean and hungry look;

He thinks too much;

Such men are dangerous.

Sheldon’s (1942) constitutional approach to psychology also attempted to relate body type and personality. Endomorphs were seen as extrovert, fun-loving and outgoing, while mesomorphs were seen as aggressive, risk-taking and athletic. Lastly, ectomorphs were seen as introverted, inhibited and shy. Although such broad generalisations may initially seem hard to sustain, more recently it has been suggested that there might be a physiological basis to this, relating to people’s relative levels of neurotransmitters and hormones such as serotonin, testosterone and dopamine.

Clearly, there are many factors which impact on the development of human personality. These factors include genetics, biology, life experiences, trauma, situational factors, workplace culture, national, gender, ageing, family, educational and perhaps even constitutional factors.

In considering why there are numerous methods and styles of assessment available and to fully understand the history of assessment, it is important to look at the theories of personality from which these methods of assessment have evolved. Over the years, there have been many attempts to define and explain personality and there are considerable variations among psychologists as to the precise definition of the term ‘personality’.

Carl Rogers (1959) defined personality as being about the self: an organised permanent subjectively perceived entity, which is at the very heart of all our experiences. Rogers was a strong believer in the notion that the way in which individuals perceive themselves influences both their behaviour and the way they see the rest of the world.

Gordon Allport is often called the father of personality theory. He was a trait theorist who believed in the individuality and uniqueness of the person, and that people have consistent personalities. His lexical approach (sometimes known as the Lexical Hypothesis) described many of the different characteristics or facets of personality. It was based on the idea that personality variables which are important to people’s lives are manifest in the language they use. Allport and Odbert (1936) engaged in research where they categorised dictionary entries into various conceptual lists to produce nearly 20,000 items which were used to describe personality and behaviours. Allport believed that each person had traits of various types.

Raymond Cattell, author of the 16PF personality questionnaire, defined personality as being something which ‘enables us to predict what a person will do in real life situations’ (Cattell, 1946). T...