How can you determine the reliability of exam questions after grading?

The reliability of exam questions can be determined by assessing the consistency of results each time the question is administered. When a question demonstrates the same results across multiple administrations, it indicates that the question accurately measures what it is intended to measure, independent of various external factors such as test-taker variability or specific cohort differences. This consistency is crucial for establishing the reliability of an assessment tool, ensuring that it can be trusted to yield similar outcomes under similar conditions, thus providing a stable measure of student understanding.

The other options focus on different aspects of scoring and performance but do not directly address the concept of reliability in the same way. For example, the equal number of high and low scoring students getting the correct answer might point to the question's fairness or difficulty but does not measure consistency over time. Similarly, the number of items students answered correctly or overall scores provide information about student performance but do not offer insights into whether the questions themselves reliably assess knowledge consistently across different scenarios.