The University of South Australia’s latest machine-learning model enables swift and cost-efficient creativity evaluations in schools, utilizing the TCT-DP test to yield rapid and accurate outcomes.
During a period in which artificial intelligence (AI) is managing repetitive, predictable tasks, promoting creativity in educational settings has become more vital than ever. While most educators acknowledge creativity as a crucial 21st-century competency, the absence of validated and dependable creativity assessments makes it challenging for schools to gauge creativity.
The University of South Australia has now crafted an innovative machine-learning model that provides access to top-notch, purpose-built creativity assessments, capable of scoring evaluations in a fraction of the time and cost.
Transforming Creativity Assessments with AI
Applied to the prevailing empirical creativity test – Test of Creative Thinking – Drawing Production (TCT-DP) – the recent algorithm assesses a test within a single millisecond, contrasting with the standard 15-minute human-evaluated test. This breakthrough could save educators countless hours in an already overstretched timetable.
The principal investigator, UniSA’s Prof David Cropley, describes the algorithm as a groundbreaking advancement for educational institutions.
“Creativity is an indispensable proficiency for upcoming generations, especially since it’s a capability that cannot be mechanized,” Prof Cropley asserts. “However, due to the absence of inexpensive and effective tools for measuring creativity in schools, students are either not being evaluated, or are being graded subjectively, leading to inconsistency and unreliability.”
Advantages of Automated Creativity Assessment
Prof Cropley elaborated: “The TCT-DP test has long been recognized as the foremost instrument for evaluating creativity in school-age children, but due to its high cost, slow pace, and labor-intensive nature, it remains out of reach for most educational institutions.
“Our algorithm transforms this scenario. Not only does it drastically reduce the expenses of running the algorithm by more than 20-fold, but the results are swift and remarkably precise.
“For instance, manually grading a test for a school with 1000 students would entail an expenditure of roughly $25,000 and necessitate around 10 weeks to receive the results; with UniSA’s algorithm, the same evaluation could be carried out for approximately $1000 with outcomes delivered within 1-2 days.
“This makes the assessment accessible to schools and educators, providing them with the means to evaluate creativity accurately and affordably.”
Spotting Varied Student Talents
Co-researcher, UniSA’s Dr. Rebecca Marrone, highlights the benefits of being able to evaluate and quantify creativity, especially for students who might be disregarded at times.
“Assessing creativity opens up a pathway beyond conventional intelligence assessments,” Dr. Marrone notes.
“Evaluating creativity aids in identifying students who may possess talents that aren’t evident through conventional testing methods in schools. For instance, a student who performs poorly on standard IQ tests but exhibits high creativity might easily go unnoticed.
“Nurturing creativity also safeguards children on the lower end of the achievement spectrum by honing a skill that isn’t susceptible to automation, helping shield them from the impacts of digital transformation.”
Rollout and Future Outlook
The algorithm is presently in the development phase as a desktop application for educators to use in classrooms. Ahead of this, classroom teachers keen on utilizing the TCT-DP are encouraged to engage with the UniSA team to discuss their requirements.
Image Source: Poca Wander Stock / Shutterstock
