Advanced statistical analyses including structural equation modeling, cluster analysis, and multinomial logistic regression provide education researchers with additional methods for testing hypotheses about binary, ordinal, and continuous data. This poster will present results from three studies conducted by the ACS Examinations Institute. 1.) The use of structural equation modeling to gage the impact of objective and subjective exam item complexity on student performance. 2.) The use of cluster analysis to identify levels of chemistry faculty familiarity with assessment terminology. 3.) The use of logistic regression to understand the differences on several assessment instruments between students enrolled in several different general chemistry courses. We will present a brief background, the types of questions that can be answered, and how to interpret the results for each method.

Using advanced statistical analyses in chemical education research: Applications of structural equation modeling, cluster analysis, and multinomial logistic regression

245th National Meeting (2013)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.