{
  "id": "1506.00852",
  "version": "v1",
  "published": "2015-06-02T12:03:30.000Z",
  "updated": "2015-06-02T12:03:30.000Z",
  "title": "Peer Grading in a Course on Algorithms and Data Structures: Machine Learning Algorithms do not Improve over Simple Baselines",
  "authors": [
    "Mehdi S. M. Sajjadi",
    "Morteza Alamgir",
    "Ulrike von Luxburg"
  ],
  "comment": "Workshop on Machine Learning for Education, International Conference of Machine Learning (ICML), 2015",
  "categories": [
    "cs.LG",
    "stat.ML"
  ],
  "abstract": "We used peer grading in a course on algorithms and data structures at the University of Hamburg. During the whole semester, students repeatedly handed in solutions to exercises, which were then evaluated both by teaching assistants and by peer grading. We tried different methods from the machine learning literature to aggregate the peer grades in order to come up with accurate final grades for the submitted solutions (supervised and unsupervised, methods based on numeric scores and ordinal rankings). We found that none of them improves over the baseline of using the mean peer grade as the final grade.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2015-06-02T12:03:30.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "machine learning algorithms",
      "peer grading",
      "data structures",
      "simple baselines",
      "mean peer grade"
    ],
    "tags": [
      "conference paper"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}