Analysis different between Classical Test Theory dan Modern Test Theory
Instructions for Completing the Student Activity Sheet
- The Student Activity Sheet must be completed individually.
- The Student Activity Sheet must be completed before the student attends the lecture.
- Students are not allowed to attend the lecture if they have not completed the Student Activity Sheet.
Introduction
Once upon a time, Krishna wanted to test five individuals in archery. Four targets were prepared based on their level of difficulty: a tree, a deer, a bird, and a thread. Of the four targets, Gatotkaca was only able to shoot the tree and the deer. Krishna saw no problem with that—that was Gatotkaca’s level of competence. Gatotkaca was then allowed to go home, unlike the remaining four. The other four were gathered and questioned one by one.
Petruk could not shoot the tree or the deer, but he could hit the thread. When Krishna turned his eyes toward him, Petruk confessed, "When I shot the thread, I used a different arrow that I had enchanted." Petruk had cheated. Then Krishna asked Srikandi why she couldn’t shoot the deer, even though she could shoot the tree, bird, and thread. Srikandi answered, "I was sleepy, dear brother." So Srikandi had been drowsy when shooting at the deer. Krishna then asked Abimanyu why he was able to shoot the thread, even though he had failed to hit the deer and bird. It turned out Abimanyu had guessed, and his success in hitting the thread was just a coincidence. Lastly, Krishna looked at Arjuna. Arjuna said, "I didn’t want to shoot the tree, because it would undermine my competence as a true archer."
By using different levels of difficulty, we can observe the psychological dynamics behind individual responses. In addition, we can also predict scores based on missing data (intentionally missing), as in Arjuna’s case.
From this case, it can be stated that in the context of social science measurement, errors are inevitable. Instruments in social science measurement are prone to bias, due to psychological factors such as those illustrated earlier. The Conventional Test Theory (CTT), which is often used, is not capable of covering these limitations—hence the emergence of Item Response Theory (IRT). IRT is a modern theory that not only focuses on the test but also considers the test-taker. IRT can predict errors in measurement, including predicting missing scores.
As an introduction, please watch the following video to observe the differences between CTT and IRT: https://youtu.be/5nimz8A_VDM
