Theses and Dissertations at Montana State University (MSU)
Permanent URI for this collectionhttps://scholarworks.montana.edu/handle/1/733
Browse
1 results
Search Results
Item Examining the psychometric functionality of the force concept inventory(Montana State University - Bozeman, College of Letters & Science, 2020) Eaton, Philip Dale; Chairperson, Graduate Committee: Shannon Willoughby; Keith Johnson and Shannon Willoughby were co-authors of the article, 'Generating a growth-oriented partial credit grading model for the force concept inventory' in the journal 'Physical review physics education research' which is contained within this dissertation.; Barrett Frank and Shannon Willoughby were co-authors of the article, 'Examining the effects of item chaining in the force concept inventory and the force and motion conceptual evaluation using local item dependence' submitted to the journal 'Physical review physics education research' which is contained within this dissertation.; Shannon Willoughby was a co-author of the article, 'Confirmatory factor analysis applied to the force concept inventory' in the journal 'Physical review physics education' which is contained within this dissertation.; Shannon Willoughby was a co-author of the article, 'Identifying a preinstruction to postinstruction model for the force concept inventory within a multitrait item response theory framework' in the journal 'Physical review physics education' which is contained within this paper.To improve the current understanding of the Force Concept Inventory (FCI), both a response-option-level analysis and a dimensionality analysis were proposed and applied. The response-option-level analysis used polytomous item response theory to reveal that the response options on the FCI are generally functioning appropriately, with two questions being identified as likely malfunctioning. To address the question of the FCI's dimensionality, an analysis of local item independence using item response theory was proposed and performed. Results indicate that the FCI is a multi-factor instrument, not a unidimensional instrument as it is often assumed. As a result of this analysis, three factor models were proposed and tested using confirmatory factor analysis and confirmatory multi-trait item response theory. All of these models were found to adequately explain the factor structure of the FCI within each of the statistical frameworks. The results from these investigations can be used as a starting point for further analysis and directing future improvements of the FCI.