From: A review on process-oriented approaches for analyzing novice solutions to programming problems
Study | What has been done | PL | Tools | Gran. |
---|---|---|---|---|
Used the Blackbox dataset to analyze the frequency, time-to-fix, and spread of errors among users | Java | BlueJ | C | |
Detected high- and low-performing students after the very first week of the programming course to provide better support for them using EQ, WS, and standard machine learning techniques | Java | Test My Code | KS | |
Proposed and derived NPSM to predict students programming performance, and compared its explanatory power against EQ and WS | C++ | OSBIDE | All | |
Evaluated the performance of EQ in multiple contexts (different datasets, different languages, working practices, and student backgrounds) | C, Python, Java | CloudCoder, PCRS, Test My Code | S | |
Attempted to predict accurately failing and passing students in the middle of the semester using a decision tree classifier | – | PASTA (auto. marking and feedback sys.) | S | |
Aimed to automatically analyze students’ solving process in programming exercises using visualization | Python | Thonny | All | |
Correlated keystroke latency with exam performance and programming experience | Java | Test My Code | KS | |
Introduced RED to quantify errors, and compared it with EQ | Java | Custom Java editor | C |