What's Wrong with Psychometric Tests?
Psychometric testing has been with us for some time and maintains a wide variety of applications. For example, many larger organisations now use psychometric testing for graduate recruitment alongside promotion and performance management processes.
Importantly, the prevailing view is that psychometric testing helps us to objectively identify the best talent whilst showing support for concepts such as meritocracy and fairness. And indeed, smaller organisations are also looking to introduce psychometric testing as means of finding suitable candidates with limited resources.
However, behind a veil of objectivity, lie significant challenges for psychometric tests and whether they can reliably be used to predict on-the-job performance or indeed intelligence.
Where Did Our Passion for Psychometric Testing Start?
Psychometric testing has a long history. One of the earliest examples was a test used in World War 1 to identify US military personnel at risk of shell shock. It involves answering yes/no to 116 questions and you can find it here if you'd like to take it. Several tests would later be developed with both civilian and military applications including the testing of recruits (Zibby and Zickar, 2008). Since that time, their use has proliferated to include applications across areas such as personality, cognitive ability and skills-based assessments.
Importantly, moreover, what these developments have achieved is to advance an entire paradigm which assumes that knowledge can be objective not just in nature but in the social sciences too - that optimal skills, intelligence and best practices do exist as objective truths.
Positivism
Psychometric tests assume that the individual and their intelligence, personality and skills can be measured and measured accurately. In this sense, psychometric tests are seen as instruments which measure naturally occurring phenomena that occur irrespective of the measurement used or the person (or platform) used to conduct the measurement. To use psychometric tests is to assume first of all that a candidate's intelligence, skills and abilities can be objectively measured (Positivist) regardless of their experience of reality at that time (Constructivist).
Constructivism
Unlike Positivism, Constructivism holds that the an objective reality does not exist at least in regards to the social sciences. Rather, reality is constructed and is unique to the individual experiencing or witnessing a phenomenon. For example, a candidate who takes a psychometric test constructs their reality through responses to the questions on the paper. Those questions could embody how a candidate anticipates her future self, how she feels currently based on the experiences and interactions she's had that day. Past assessments could also have a bearing on how her reality is constructed whether positive or negative in addition to broader cultural norms and gender stereotypes.
All of these factors (and many more besides) helps us to question whether for any individual there exists an objective reality which can neatly and reliably encapsulate a specific trait.
What This Means in Practice?
Psychometric testing promises us knowledge. Knowledge about a candidate's current abilities and future, on-the-job performance. They are meant to be objective in measuring those attributes which are most relevant for a particular role or perhaps whether that candidate will be able to 'fit' into the team. Some considerations and links to thought provoking articles:
How well does the test predict on-the-job performance?
How well do specific cognitive abilities predict on-the-job performance?
What other factors predict strong on-the-job performance and how significant are those factors?
How well defined is the job itself?
Who decides what skills or abilities are most relevant? Are those skills and abilities gendered or neutral?
What is the organisation's dominant culture? Where might 'fit' lead to exclusion?
And one step further:
Does a psychometric test unfairly favour able-bodied individuals?
How can priming different individuals in different ways affect test scores?
Concluding Remarks
Psychometric testing is here to stay and many organisations will continue to use a variety of assessments in order to (hopefully) identify suitable candidates. What this article has sort to do is shed some light on a paradigm (Constructivism) which challenges this view thereby helping us to identify some of the ways in which we can better conceptualise how and where psychometrics can be used responsibly.
References:
Gibby, R. E., & Zickar, M. J. (2008). A history of the early days of personality testing in American industry: An obsession with adjustment. History of psychology, 11(3), 164.