In statistics, internal consistency is a reliability measurement in which items on a test are correlated in order to determine how well they measure the same construct or concept. Reliability shows how consistent a test or measurement is; "Is it accurately measuring a concept after repeated testing?"
Internal consistency is a check to ensure all of the test items are measuring the concept they are supposed to be measuring.
For example, to test the internal consistency of a test a teacher may include two different questions that measure the same concept. If the student gets both questions correct or both wrong then the internal consistency of the test could be said to be good - the student understands the concept if they got both questions right and doesn't understand the concept if they got both wrong. If the student gets one question right and one wrong then the internal consistency is in question.