When it comes to hiring, it can be difficult for an employer to find the perfect person for the job. As a result, organizations often use personality scales to help determine if a candidate is the right fit.
Although there are several widely used personality tests on the market, organizations may be looking for traits or skills that are not measured by already existing scales. Creating a new scale—which requires the work of experts such as personality, organizational, social, or clinical psychologists—can be time-consuming and expensive.
With that in mind, Ivan Hernandez, an assistant professor in the psychology department at Virginia Tech, wanted to find a way to make creating personality scales easier and more accessible.
“As psychologists, there are so many different aspects of personality that we would be interested in measuring,” Hernandez said. “But the hard part is how to do it? How do you find the right questions to find out if a person is a good friend, to find out if a person would be a hard worker, to find out if a person is emotionally intelligent?”
While these questions are usually crafted by subject matter experts, Hernandez suggested an alternative source: artificial intelligence.
Together with research consultant Weiwen Nie of Hogan Assessment Systems Inc. Hernandez created a framework for using various natural language processing models to help researchers develop valid psychological scales.
In the traditional method of creating personality scales, subject matter experts are asked to create a set of items that may correspond to a particular personality trait, for example, “I like going to parties” as a measure of ‘extroversion. This pool forms the basis for building the scale, which is administered and tested before being deployed.
In Hernandez’s framework, a transformer-based language model generates the artificial intelligence-based (AI-IP) element pool, consisting of a million new elements, far more than any pool of elements. experts could create. Additional language processing models narrow the pool to those items most relevant to the desired construct, such as extraversion.
Essentially, this multi-model framework allows researchers to create longer, consistent scales from a small set of relevant items.
The best part? Scales created using AI-IP perform just as well as scales created through the human process of validation and calibration.
“When we give these objects to people and show them real objects that weren’t made by a computer, people just can’t tell the difference,” Hernandez said. “That song and dance we do as humans to create custom scales through committee can really be solved by leveraging the internalized knowledge of an AI model.”
The framework designed by Hernandez and Nie can help organizations reduce the time and expense involved in creating personality scales. By relying on artificial intelligence to create the pool of elements, the subjectivity, inconsistency and biases inherent in humans are also circumvented.
More importantly, the framework, which is freely available, achieves Hernandez’s goals of improving accessibility to personality measures. Now anyone, whether it’s a lawyer wanting to assess the reliability of a jury or a college student questioning the cleanliness of their new roommate, can create a personality scale.
“This framework was meant for organizational use, but I really think it’s something that can help ordinary people,” Hernandez said. “People are interested in personality, but I think they may not know of more validated ways to explore their interest than the metrics most commonly seen on social media or in pop culture.”
An article describing the framework and how it was created is being published in a 2023 Staff psychology special issue focusing on artificial intelligence and machine learning applications in staff selection and staffing.
Those interested in experimenting with the AI-IP generation tool can access the application online.
Ivan Hernandez et al, AI-IP: Minimizing the Guesswork of Personality Scale Item Development Through Artificial Intelligence, Staff psychology (2022). DOI: 10.1111/peps.12543
Provided by Virginia Tech
Quote: Using Artificial Intelligence to Gain Insights into Personality (2023, January 5) Retrieved January 5, 2023 from https://phys.org/news/2023-01-artificial-intelligence-gain-insights-personality.html
This document is subject to copyright. Except for fair use for purposes of private study or research, no part may be reproduced without written permission. The content is provided for information only.