Error message

Notice: Only variables should be passed by reference in theme_biblio_tabular() (line 285 of /var/www/beyondmultiplechoice/sites/all/modules/biblio/includes/biblio_theme.inc).

Using lexical analysis software to understand student knowledge transfer between chemistry and biology

TitleUsing lexical analysis software to understand student knowledge transfer between chemistry and biology
Publication TypeConference Proceedings
Year of Publication2009
AuthorsHaudek, KC, Moscarella, RA, Urban-Lurain, M, Merrill, JE, Sweeder, R, Richmond, G
Conference NameNational Association of Research in Science Teaching Annual Conference
Date PublishedApril 17-21
PublisherNational Assocation for Research in Science Teaching
Conference LocationGarden Grove, CA
KeywordsAACR, Biology, chemistry, Lexical analysis, Transfer
Full Text

Time and resources often prevent using constructed response assessments in large undergraduate science. We investigate the utility of using lexical analysis software to categorize student responses and uncover undergraduate student misconceptions in chemistry and biology. Students were randomly assigned a question set consisting of two questions relating to a single topic. Student responses were analyzed using SPSS Text Analysis for Surveys, using a custom library of science-related terms. The resulting analyses of student responses suggest potential barriers and connections between students understanding of these topics. Only 38 out of 160 students linked reaction spontaneity with thermodynamics. Student responses for one question set were rated using a scoring rubric by two independent scorers. Analysis of an acid/base question set showed student deficiencies in predicting pH behavior of functional groups in biology. After this scoring was complete, discriminant analysis was used to create classification functions that could predict human expert scores with 65.4% and 82.4% accuracy (p < .000). This study suggests that computerized lexical analysis may be useful for automatically categorizing large numbers of student open-ended responses.

Attachments: 

thumbnail of small NSF logo in color without shading

This material is based upon work supported by the National Science Foundation (DUE grants: 1438739, 1323162, 1347740, 0736952 and 1022653). Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the NSF.