ERIC Number: ED492946
Record Type: Non-Journal
Publication Date: 2006-Apr-21
Concept Map Scoring: Empirical Support for a Truncated Joint Poisson and Conway-Maxwell-Poisson Distribution Method
Allen, Bradford D.
Online Submission, Paper presented at the Annual Meeting of the New England Mathematical Association of Two Year Colleges (32nd, Manchester, NH, Apr 21, 2006)
Concept map structure, testing, and scoring methods are discussed and a new scoring methodology is introduced using the breadth and depth of individual concept maps. The scoring method proposed here provides advantages of grading "on a curve" such as the ability to estimate and compare the complexity of different concept maps, the ability to measure the statistical distance between individual map scores, the ability to define map-score quantiles, a reduction in grading bias, and an overall improvement in the reliability and validity of concept map evaluation. Empirical support for underlying probability distributions is presented and mathematical aspects are discussed. Tables of concept map score distributions for one branch and one node are provided and may be used to very quickly assign concept map scores based on one branch and one node. (Contains 7 tables and 3 figures.) [The research for this paper was funded in part by a grant from the Academic Information Technology Committee for Academic Initiatives, Florida Institute of Technology, Fall 2003.]
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Sponsor: Florida Inst. of Tech., Melbourne.
Authoring Institution: N/A