View in EDS

Using Lexical Semantic Techniques To Classify Free-Responses.

Saved in:
Bibliographic Details
Title: Using Lexical Semantic Techniques To Classify Free-Responses.
Language: English
Authors: Burstein, Jill, Wolff, Susanne, Lu, Chi
Peer Reviewed: N
Page Count: 19
Publication Date: 2001
Document Type: Reports - Evaluative
Descriptors: Essays, Performance Based Assessment, Scoring, Test Items, Training
Abstract: The research described in this paper shows the use of lexical semantic techniques for automated scoring of short-answer and essay responses from performance-based test items. Researchers used lexical semantic techniques in order to identify the meaningful content of free-text responses for small data sets. One data set involved 172 training responses in a formulating hypotheses situation, with 32 categories. A second set contained 200 training responses and 228 test items, also of the formulating hypotheses type. The third set consisted of 200 essays from a paper-and-pencil test that had been rated "excellent" by human raters. The research demonstrated applications of lexical semantic techniques for free-text responses of varying length and in different subject domains. Prototype designs and the results of the different prototype applications are discussed. (Contains 5 tables and 23 references.) (SLD)
Entry Date: 2002
Accession Number: ED458268
Database: ERIC
Description
Abstract:The research described in this paper shows the use of lexical semantic techniques for automated scoring of short-answer and essay responses from performance-based test items. Researchers used lexical semantic techniques in order to identify the meaningful content of free-text responses for small data sets. One data set involved 172 training responses in a formulating hypotheses situation, with 32 categories. A second set contained 200 training responses and 228 test items, also of the formulating hypotheses type. The third set consisted of 200 essays from a paper-and-pencil test that had been rated "excellent" by human raters. The research demonstrated applications of lexical semantic techniques for free-text responses of varying length and in different subject domains. Prototype designs and the results of the different prototype applications are discussed. (Contains 5 tables and 23 references.) (SLD)