An IOP Introduction to Artificial Intelligence

Artificial Intelligence (AI) is not only here to stay, but will also revolutionize how IOPs practice their discipline. In today’s article, we’d like to take you through what AI is (and is not), as well as some powerful practical applications of the technology in the field of talent management and selection.

In addition, we will show you how TTS’s best-of-breed AI partner, HireVue, has harnessed the power of AI in their interviewing and assessment solutions.

What are AI-assisted assessments?

Although a seemingly simple matter, it is perhaps telling how seldom this question is asked in discussions of AI and IOP.

A very simple working definition of AI is that it is a technology that allows machines or computers to appear to have human-like intelligence.

How is this accomplished?

At its core, AI relies on algorithms: Decision rules that allow the completion of a series of tasks based on simple rules and principles.

This capacity to learn from data, as well as the cumulative effect such learning has on eventual predictive accuracy, means that AI can be a game-changer in helping IOPs reach more valid and accurate conclusions based on assessment data.

What will AI mean for IOPs?

Despite fears to the contrary, the actual application of AI in the discipline of IO Psychology is unlikely to replace psychologists, recruiters or other talent management professionals.

Instead, Artificial Intelligence will become an efficiency tool and competitive advantage for those who choose to use it. In the words of Dr Nathan Mondragon, the Chief Industrial and Organizational Psychologist at TTS assessment partner, HireVue, AI will:

  • Help recruiters and talent acquisition teams assess more candidates more quickly and accurately.
  • Be akin to putting more reviewers to work for you without adding more people to our team.
  • Democratise the hiring process by minimising bias in selection decisions.

In addition, applications such as HireVue’s AI-assisted virtual interviewing allow recruiters and IOPs alike to feel confident that the early-stage pool of candidates that was filtered using AI algorithms is highly likely to be successful at the job, thus eliminating potential errors in selection decision-making early in the hiring process. Think of the candidate that are shortlisted only based on the University or Schools they attended and all the strong candidates that will be rejected because they attended the “wrong” school.

One of the main reasons for this confidence is that HireVue’s AI algorithms are ultimately based on and intended to mimic the reasoning of experts (in the case of virtual interviewing, the reasoning and judgment of expert raters of candidates and their interview answers).

This is an important point to consider. HireVue’s AI-assisted assessments do not use generic or abstracted data mining to arrive at predictions about candidates’ potential. Instead, the algorithms are “trained” through feedback from actual human experts and thousands of rated interviews.

It is well-known that human raters are limited in terms of biases. Countless such cognitive distortions have been documented and include halo effects, attractiveness biases, in-group biases and the fundamental attributional error.

A powerful feature of using AI-assisted rating is that adverse impact can be monitored and mitigated through the adaptation of the algorithms used. Bias can therefore be substantially reduced and effectively managed.

An example will best illustrate this principle:

Using AI-assisted rating of virtual interviews for instance, HireVue data scientists can analyze whether any of the protections used by IOPs to ensure fair treatment are being violated, such as the 4/5ths Rule, or significant statistical differences between sub-groups.

Based on such research, the specific data points responsible for such differences (and the consequent adverse impact) are then identified and reduced or eliminated from the algorithm. Therefore, bias mitigation is a key strength and competitive advantage of using AI-assisted rating of video assessments.

Final Thoughts

The goals of Artificial Intelligence-assisted video assessments are to:

  • Help companies reduce time to hire and increase retention.
  • Help talent professionals in selecting the best candidates for the needs of the job by screening more candidates in a fast, fair, consistent fashion.
  • Minimise any algorithmic or human bias from the hiring process.
  • Provide an excellent candidate experience

HireVue’s suite of assessments includes video interviewing, game-based assessments, and coding challenges that are all integrated in the final hiring recommendation.

In this article, we discussed how Artificial Intelligence powers can be a more accurate, less biased, more engaging screening process for both recruiters and candidates if it is designed and deployed with the necessary behavioral scientific insights. Indeed, a cornerstone of HireVue’s method is to gather candidate experience ratings, which have been consistently above the industry norm. In other words, using HireVue has a positive value proposition effect over and above the accuracy and efficiency benefits such assessments offer.  All of this is backed by a robust data analysis and behavioral scientific underpinnings.

If you are interested in how HireVue AI-assisted assessments may help your talent or recruitment function, why not drop us al line at info@tts-talent.com

Sources:

Mondragon, N. & Slifka, J. (2019). Creating AI-Driven Pre-Employment Assessments. HireVue online article.