Premium
Using Rasch theory to guide the practice of survey development and survey data analysis in science education and to inform science reform efforts: An exemplar utilizing STEBI self‐efficacy data
Author(s) -
Boone William J.,
Townsend J. Scott,
Staver John
Publication year - 2011
Publication title -
science education
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.209
H-Index - 115
eISSN - 1098-237X
pISSN - 0036-8326
DOI - 10.1002/sce.20413
Subject(s) - rasch model , set (abstract data type) , science education , item response theory , psychology , mathematics education , computer science , data science , philosophy of science , quality (philosophy) , management science , psychometrics , engineering , epistemology , clinical psychology , developmental psychology , programming language , philosophy
Over many decades, science education researchers have developed, validated, and used a wide range of attitudinal instruments. Data from such instruments have been analyzed, results have been published, and public policies have been influenced. Unfortunately, most science education instruments are not developed using a guiding theoretical measurement framework. Moreover, ordinal‐level attitudinal data are routinely analyzed as if these data are equal interval, thereby violating requirements of parametric tests. This paper outlines how researchers can use Rasch analysis to develop higher quality science education instruments and why researchers must use Rasch analysis to prepare appropriate ordinal‐level data for parametric analyses. Detailed recommendations are set forth to help science education researchers increase the rigor of attitudinal instrument development and analysis. Suggestions are also set forth regarding how philosophical and analytical Rasch techniques align with broad goals of the science education research community. © 2010 Wiley Periodicals, Inc. Sci Ed 95: 258–280, 2011