Schools Experiment With AI but Struggle to Measure Its Effect on Students
By Emma Carter
Image information
This is a representative image for this Daily News lesson.
Intro
Schools are experimenting with artificial intelligence, yet many are finding it difficult to determine whether the technology benefits students or harms their learning.
Vocabulary
| Word | Part of speech | Meaning | Example |
|---|---|---|---|
| fundamental adjective | adjective | Meaning: basic and essential to understanding or addressing an issue | Example: Reliable evidence is fundamental to responsible educational policy. |
| influence noun | noun | Meaning: the power or effect that something has on a situation or outcome | Example: Parental expectations can exert a considerable influence on academic motivation. |
| eventual adjective | adjective | Meaning: occurring after some time or following a series of developments | Example: The eventual outcome depended on several difficult decisions. |
| Conversely adverb | adverb | Meaning: in contrast with what has just been stated | Example: A policy may reduce costs; conversely, it may undermine long-term quality. |
| distinguish verb | verb | Meaning: to recognize or identify the difference between things | Example: Researchers must distinguish correlation from genuine causation. |
| credible adjective | adjective | Meaning: believable and supported by reliable evidence | Example: The proposal requires a credible account of how the funds will be used. |
| consequences noun | noun | Meaning: the results or effects that follow from an action or decision | Example: Before adopting the policy, officials considered its unintended consequences. |
The story
Schools are experimenting with artificial intelligence, but many are encountering a fundamental difficulty: they cannot easily measure whether the technology is improving students’ education or producing negative effects.
The challenge is more complicated than simply asking whether a digital tool appears useful. An experiment may introduce AI into an existing classroom, assignment, or school process, yet its influence can be difficult to separate from the many other factors that shape learning. Students’ experiences, teachers’ decisions, and the design of the activity may all affect the eventual outcome.
Measurement also requires a clear definition of success. A tool might save time, offer rapid assistance, or make a task feel more accessible. Those qualities could be valuable, but they do not automatically demonstrate that students are learning more effectively. Conversely, a result that seems disappointing at first may not reveal the full impact of a new approach.
This creates a significant problem for schools that are still testing how AI should be used. If they cannot evaluate the results with confidence, they may struggle to distinguish a genuinely beneficial experiment from one that merely appears promising. They may also find it harder to decide whether a particular use of the technology should be expanded, revised, or abandoned.
The issue reflects a broader tension surrounding innovation in education. New tools often arrive before institutions have developed agreed methods for assessing them. Schools therefore face two tasks at once: exploring what artificial intelligence might contribute and establishing credible ways to judge its consequences for students.
For now, the central question is not simply whether schools can introduce AI. It is whether they can determine, with enough clarity to guide future decisions, what the technology is actually doing for learners. That uncertainty is shared by many schools conducting experiments, making evaluation as important as adoption.
Warm-up questions
- What makes it difficult to determine whether a new educational technology genuinely improves learning?
- Which outcomes should schools measure when evaluating AI: efficiency, achievement, accessibility, or something else?
Comprehension questions
- What fundamental difficulty are schools encountering as they experiment with AI?
- Why is it difficult to isolate AI’s influence on learning outcomes?
- Why does saving time not necessarily prove that students are learning more effectively?
- What decisions might schools find harder to make without confident evaluation?
- What two tasks must schools undertake simultaneously?
Opinion questions
- Should schools prioritize measurable academic gains over less tangible benefits such as accessibility or reduced workload? Explain your position.
- How much evidence should schools require before expanding an experimental use of AI?
Further discussion
- What kinds of research designs could help schools separate AI’s effects from teachers’ decisions and students’ prior experiences?
- How might schools establish shared evaluation standards without discouraging experimentation and innovation?
Writing prompt
- Write a balanced analytical essay explaining how schools could evaluate AI experiments, addressing both measurable outcomes and unintended consequences.
