Marking concept maps by hand is slow, subjective, and the reason plenty of teachers quietly stop assigning them. The Kit Build approach sidesteps part of that. Because students rebuild a map from a fixed set of concepts and links, software can compare the result against the expert map and score it without a person reading every proposition.
A 2023 study put that automatic scoring to the test. The authors asked two things: whether better maps, as the software measures them, go together with stronger higher order thinking, and whether machine scores agree with the scores a human assessor produces.
Assessing a concept map means judging structure, accuracy, and completeness at once. It is defensible work, but it does not scale. A stack of thirty maps can eat an evening, and by the time the marks come back the class has moved to another topic.
The automatic scores held up against manual assessment closely enough that the authors describe the method as promising. Better map quality also went hand in hand with better answers on higher order questions, which is the outcome the assessment is meant to predict.
That combination is what makes the result useful. Scoring a map the moment it is finished gives students a signal while the material is still fresh, and it returns to teachers the hours that marking would have taken.
The study does not claim software should replace a teacher's judgement, and it does not try to score the ideas a student brings that were never in the expert map. What it shows is narrower and more practical: for this kind of closed mapping task, the automatic score is good enough to act on.
Nurmaya et al. (2023). Investigation of the Relationship Between Map Quality and Higher-Order Thinking in Kit-Build Concept Map. Human Interface and the Management of Information, Lecture Notes in Computer Science.
