Image this: Tomorrow’s graduates stroll into workplaces the place AI instruments are as frequent as e-mail — diagnosing affected person signs, analyzing market developments, optimizing provide chains or designing new infrastructure. From healthcare to advertising to engineering, almost each discipline is being remodeled. Are our faculties getting ready them for this new actuality? And do we have now an efficient technique of assessing such readiness?
At Gwinnett County Public Faculties (GCPS), educators are decided to verify each solutions are “sure.” Their mission is to make sure each scholar is “AI prepared” — ready to make use of rising applied sciences, like generative AI, in an moral and accountable method in class, life and future work, no matter the place these careers take them. To help this purpose, GCPS led the event of each an AI readiness framework and a companion diagnostic evaluation.
In 2019, GCPS, in collaboration with a number of companions, created an AI readiness framework that focuses on six core areas: Information Science, Mathematical Reasoning, Inventive Drawback Fixing, Ethics, Utilized Experiences and Programming. The framework was developed with enter from district material specialists (together with pc science, math and science academics) and exterior companions.
As a way to assist make the framework informative and actionable, the district partnered with the ISTE analysis group in 2025 to develop a diagnostic evaluation device that measures scholar AI readiness throughout choose expertise outlined within the framework. Diagnostic assessments, versus summative assessments, measure college students’ present information and expertise, serving to educators determine gaps and areas for progress, and information academics and college leaders towards the place college students would possibly want extra instruction, assets or help to satisfy studying outcomes.
A Systematic Method to Check Design
Right here’s how the district and the analysis group introduced the AI readiness diagnostic evaluation to life:
Defining aims and creating the framework
The group needed to account for sensible concerns: Who would take the take a look at? How wouldn’t it be delivered? What time constraints existed?
Whereas the AI readiness framework covers preK-12, the group started by designing a diagnostic for highschool college students in grades 9-12. They knew the evaluation wanted to be digital (to maximise flexibility) and fast, ideally 10 to fifteen minutes. These elements influenced the forms of questions used. To help automated scoring, the group included multiple-choice and Likert scale questions.
Creating draft questions
First, the ISTE analysis group and GCPS companions collaborated to determine framework constructs they wished to measure inside every of the six core areas. This ensured constant protection throughout all areas.
As soon as the constructs have been outlined, the group labored with material specialists — each district educators and exterior specialists in AI and schooling — to draft three to 5 objects for every assemble aligned with their experience.
Reviewing and revising
After drafting the objects, the analysis group reviewed them for consistency and ensured that every measured just one ability. By way of the refinement course of, they narrowed the set to 2 objects per assemble throughout 26 constructs complete, creating two variations of the pilot evaluation. The varsity district then constructed the pilot assessments of their survey platform, Qualtrics, for ease of distribution.
Placing the pilot to the take a look at
College students from Seckinger Excessive College — about 1,200 complete — participated within the pilot. They have been cut up into two teams alphabetically by final identify to judge the 2 “parallel” units of things. The district confirmed that the 2 teams had comparable demographics. College students accomplished the pilot throughout their homeroom interval.
Analyzing the outcomes
Whereas knowledgeable enter ensured sturdy assemble validity, there was nonetheless a necessity to judge the reliability of each the objects and the general take a look at. The analysis group carried out a collection of psychometric analyses, together with take a look at reliability, empirical merchandise evaluation and merchandise response evaluation. These analyses helped determine which objects carried out properly and which wanted refinement or elimination.
Earlier than evaluation, the analysis group cleaned the information to eradicate questionable response patterns, corresponding to college students who accomplished the evaluation unusually rapidly and certain didn’t rigorously learn the objects.
The place This Work Is Headed
With the merchandise and take a look at analyses in hand, the analysis group and college district collaborated to provide a ultimate model of the diagnostic evaluation designed for highschool college students. They’re now exploring methods to adapt the device for different grade ranges and to include extra advanced objects, corresponding to performance-based duties that enable college students to show their expertise in real-world contexts.
Shifting ahead, the district hopes the outcomes from this diagnostic will contribute to a extra complete image of a scholar’s AI readiness, alongside different information factors like instructor evaluations, pc science coursework and capstone tasks. These mixed learnings will inform curriculum growth and scholar help methods throughout the district.
Reflections
Diagnostic measures of AI readiness can present districts with essential information for strategic planning and useful resource allocation, making certain college students are ready for a world saturated with AI. The collaboration between district leaders and the analysis group demonstrates the significance of considerate design and rigorous evaluation practices. GCPS and ISTE+ASCD hope their work can function a mannequin for different districts getting ready college students for a future with generative AI.
