Grading Rubric - spzwart/AMUSE-Tutorial GitHub Wiki
Rubric for the assessment the report of Modeling and Simulation in Astrophysics
Each of the following seven subcategories will be graded with a score between 0 and 10 based on the objectives. These numbers will be added, with the scores for Method, Discussion & conclusion, and Code weighed double, for a total score between 0 and 100. A score of 55 points or more is a passing score. Your final mark will be your score divided by 10, rounded to the nearest half point (except for scores of 55-57, which will be rounded to a 6).
| Category | Subcategory | Objective | Insufficient (<5.5) | Outstanding (>9) |
|---|---|---|---|---|
| Report | Method | Creativity | Stick to known methods | Develop novel method |
| Valid approach | You miss vital aspects of the problem | You cover all vital aspects of the problem | ||
| Results | Clarity of presentation | It is difficult to interpret your results/plots | Your results are easily interpreted and plots are clear | |
| Discussion & conclusion | Numerical effects | You do not consider the limitations/assumptions/initial conditions of your investigation. | You consider the limitations/assumptions/initial conditions of your investigation. You tie results to past research and consider its implications in astronomy. | |
| Validity of conclusions | Your conclusions are not supported by your results | Your results conclusively demonstrate your results | ||
| Git | Git history | Frequency of git commits | You only commit at the last moment | You commit regularly |
| Use of features | You do not use branches, and only upload files | You do proper commits and use branches | ||
| Code | Readability | Variable/function names are vague | Variable/function names are informative | |
| Structure | Single block of code | Many self-contained, concise functions/classes | ||
| Documentation | No comments, README | Informative comments and README | ||
| Redundant/messy code | Unused functions, commented out code | All code used, no commented-out blocks | ||
| Split over files | Single file | Files for initial conditions, runs, analysis, plots... | ||
| Rerunning code | Easy to run | No instructions on how to run | Clear walkthrough of scripts | |
| Code works | Code fails at some point | Code runs to the end | ||
| Project presentation | Audience doesn't understand what you did and what your results are | Audience perfectly knows what you did and what your results are. Good discussion. |