A district, a diagnosis and a desire to prepare for AI

Imagine this: the graduates of tomorrow come into workplaces where AI tools are as common as electronic mail – diagnose patient symptoms, analyze market trends, optimize supply chains or design new infrastructures. From health care to marketing through engineering, almost all areas are transformed. Do our schools prepare them for this new reality? And do we have an effective method to assess such preparation?

HAS Gwinnett County Public Schools (GCPS), educators are determined to ensure that the two responses are “yes”. Their mission is to ensure that each student is “Ai ready“- Prepared to use emerging technologies, as a generative AI, in an ethical and responsible way in school, life and future work, regardless of the place of careers. To support this objective, the GCPs led to the development of an AI preparation framework and an additional diagnostic assessment.

In 2019, GCPS, in collaboration with several partners, Created an AI preparation framework which focuses on six main areas: data science, mathematical reasoning, creative problems of problems, ethics, applied experiences and programming. The framework was developed with the contribution of district experts (including computer science, mathematics and science teachers) and external partners.

In order to help make the framework informative and usable, the district has teamed up Iste research team In 2025 to develop a diagnostic assessment tool that measures AI preparation for students in certain skills described in the context. Diagnostic assessments, as opposed to summative assessments, measure the current knowledge and skills of students, helping educators identify gaps and areas of growth, and guide teachers and school leaders to where students may need additional teaching, resources or support to achieve learning results.

A systematic approach to test design

Here is how the district and the research team gave life to the diagnostic assessment of the preparation of the AI:

Define the objectives and develop the framework

The team had to take into account practical considerations: who would pass the test? How would he be delivered? What time constraints existed?

While the AI ​​preparation framework covers the Prek-12, the team began by designing a diagnosis for secondary students from 9th to 12th year. They knew that the evaluation had to be digital (to maximize flexibility) and fast, ideally 10 to 15 minutes. These factors have influenced the types of questions used. To support automatic rating, the team has included questions on the multiple scale and Likert scale.

Creation of questions

First, the ISTE research team and the GCPS partners collaborated to identify the framework constructions they wanted to measure in each of the six main areas. This provided coherent coverage in all areas.

Once the constructions have been defined, the team worked with experts in the field – both district educators and IA and education specialists – to write three to five elements for each construction aligned with their expertise.

Examine and revise

After writing the elements, the research team examined them for consistency and ensured that everyone has measured a single competence. Thanks to the refinement process, they reduced the set to two elements by construction over 26 constructions in total, creating two versions of the pilot evaluation. The school district then built the pilot assessments of their investigation platform, quality, to facilitate distribution.

Put the pilot to the test

Students of Seckinger High School – about 1,200 in total – participated in the pilot. They were divided into two groups in an alphabetical manner by surname to assess the two sets of “parallel” elements. The district confirmed that the two groups had similar demographic data. The students finished the pilot during their room period.

Analyze the results

Although expert contributions guarantee a high validity of construction, it was still necessary to assess the reliability of the elements and the overall test. The research team carried out a series of psychometric analyzes, including the reliability of the tests, the empirical analysis of the elements and the analysis of the response to the elements. These analyzes helped to identify the well -performed elements and which required refinement or withdrawal.

Before the analysis, the research team cleaned data to eliminate questionable response models, such as students who finished the assessment unusually quickly and probably did not carefully read the elements.

Where this work is heading

With articles and test analyzes in hand, the research team and the school district have collaborated to produce a final version of the diagnostic assessment designed for secondary students. They now explore ways to adapt the tool for other levels and integrate more complex elements, such as performance -based tasks that allow students to demonstrate their skills in real contexts.

In the future, the district hopes that the results of this diagnosis will contribute to a more complete image of the preparation of AI of a student, alongside other data points such as teachers' assessments, computer courses and cornerstone projects. These combined learnings will inform the development of the curriculum and student support strategies in the district.

Reflections

The diagnostic measures for the preparation of AI can provide the districts of crucial data for strategic planning and the allocation of resources, ensuring that students are prepared for a saturated world of AI. The collaboration between district chiefs and the research team demonstrates the importance of thoughtful design and rigorous evaluation practices. GCPS and ISTE + ASCD hope that their work will be able to serve as a model for other districts preparing students for a future with a generative AI.