• DocumentCode
    2529088
  • Title

    An Automatic Scoring System for PDA Integrated Ecology Observation Worksheet

  • Author

    Su, I-Hsiang ; Hung, Pi-Hsia ; Hwang, Gwo-Jen ; Lin, Yu-Fen

  • Author_Institution
    Grad. Inst. of Meas. & Stat., Nat. Univ. of Tainan, Tainan, Taiwan
  • fYear
    2010
  • fDate
    12-16 April 2010
  • Firstpage
    198
  • Lastpage
    200
  • Abstract
    This study aims to develop an automatic scoring system for evaluating the ecology observation worksheets in mobile learning activities. The feedbacks for constructed response tasks were provided online to motivate students searching for detailed information in e-library, such that their inquiry abilities could be enhanced. Three PDA integrated outdoor observation activities and worksheets were arranged for elementary school´s ecology learning. A scoring rubric and two scoring methods, i.e. keyword pattern-matching (PM) and Latent Semantic Analysis (LSA), have been developed for rating the construct responses of worksheet. The scoring consistencies between human raters and the scoring methods were examined. Twenty-seven 5th to 6th graders from Tainan district participated in this study. The inter-rater reliability between two human raters was higher than 0.90. The correlation coefficients between keyword PM and LSA and human rating were around 0.88 and 0.74, respectively. That is, keyword PM had better performance than LSA in scoring the task specific construct responses.
  • Keywords
    computer aided instruction; data analysis; ecology; mobile computing; pattern matching; PDA integrated ecology; PDA integrated outdoor observation activities; automatic scoring system; ecology learning; ecology observation worksheet; keyword pattern-matching; latent semantic analysis; mobile learning activities; personal digital assistant; Educational technology; Environmental factors; Feedback; Humans; Pattern analysis; Psychology; Sequential analysis; Statistics; Testing; Text categorization; Latent Semantic Analysis; automatic scoring; constructed response tasks; keyword pattern-matching; scoring rubrics;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Wireless, Mobile and Ubiquitous Technologies in Education (WMUTE), 2010 6th IEEE International Conference on
  • Conference_Location
    Kaohsiung
  • Print_ISBN
    978-0-7695-3992-8
  • Electronic_ISBN
    978-1-4244-6428-9
  • Type

    conf

  • DOI
    10.1109/WMUTE.2010.33
  • Filename
    5476565