Location:
|
KFC Courts |
Date:
|
Tuesday, Aug.02 |
Time:
|
6:00PM - 6:45PM
|
Author:
|
Li Chen
School of Electronic science and engeering, Southeast University
614-292-2450, chenli.seu@163.com
|
Co-Author(s):
|
Jing Han, Chunhui Du, Yan Tu, Lei Bao
|
Abstract:
|
Both Concentration Analysis and Item Response Theory (IRT) are useful tools in education assessment. In concentration analysis, the concentration factor gives a scaled value describing how students' answers to individual questions are concentrated. Perfectly concentrated responses will produce a concentration value of 1 while random responses will produce 0. In IRT, an estimated parameter, the guessing parameter, also describes the chance of guessing in response to a question. Then it is meaningful to find out if these two factors are related. Based on the college students' FCI data collected at The Ohio State University, the concentration factor and guessing parameters for all 30 FCI questions are calculated. The results show a weak correlation between these two measures (Sig.=0.222). After comparing the algorithms, we find that concentration factor focus on all of the choices, while in IRT only the binary score (right or wrong) are used. The implications of the differences will be discussed with suggestions on revisions of the algorithms.
|
Footnotes:
|
Supported in part by NIH Award RC1RR028402 and NSF Awards DUE-0633473 and DUE-1044724
|
|
|