When the Feedback Loop Misfires
A high E-homework score followed by a disappointing exam is a symptom, not a diagnosis. This chapter examines how the digital feedback-and-practice loop can produce a correct submission without producing the understanding an independent assessment requires. The aim is to distinguish a weakness in the medium from a weakness in feedback, task design, learner use, or implementation, while keeping proposed explanations separate from what the evidence has actually measured.
The distinction matters because “the platform did not work” is often too broad to guide a decision. A student may have received useful information but not understood it. They may have understood the concept but practiced only a narrow response format. They may have found an answer without doing the intended work. Or the system may have been used long enough for participation to fade. These possibilities lead to different interpretations of the same pair of scores.
A correct answer can hide several different processes
The first failure surface is correctness without explanation. A basic digital response can tell a student that an answer is wrong, but not whether the problem came from a definition, a calculation, a choice of strategy, or a misreading of the question. For a simple item, that signal may prompt a productive check. For a difficult item, it may leave the student with no way to locate the error. The assignment score can still improve if the student keeps changing the answer, but the score then says less about conceptual repair.
This is the difference between answer correction and understanding. Understanding requires some account of why the first response failed and why the new response works. An E-homework system does not automatically supply that account. Even when it offers a written explanation or hint, the student has to connect the explanation to their own reasoning. A message can be available without being used, and it can be used without being understood.
The second failure surface is repeated trial and error. Multiple submissions are potentially valuable because they allow students to learn from mistakes and practice a repaired strategy. They can also turn the task into a search problem: change an option, submit, observe the result, and repeat until the system accepts it. That process may raise the visible assignment score while bypassing the reasoning the task was intended to develop.
The distinction is not between “good” students who revise and “bad” students who guess. The same learner may revise carefully on one task and guess on another, depending on difficulty, time, confidence, and the usefulness of the feedback. A platform that rewards only the final accepted answer makes those paths hard to distinguish. The number of attempts is therefore ambiguous. More attempts can mean more deliberate practice, more confusion, or more efficient guessing.
The third failure surface is answer access. Online homework can make it easier to find a solution, copy someone else’s response, or purchase an answer. The focal review identifies this as a risk to the authenticity of the homework process. But the supplied evidence does not quantify how often it happens, and it does not show that E-homework inevitably produces dishonesty.