This study's purpose was to investigate the dimensionality of a test containing multiple-choice (MC) and student-produced-response (SPR) item formats. Parallel tests (SPR and MC) were developed to assess skills for Florida's 1994 High School Competency Test (HSCT). Each test contained nine items selected or prepared parallel to those in the contrasting test and corresponding to a HSCT skill. Both SPR and MC tests were administered to 556 tenth graders enrolled in Florida high schools. / Five analyses attempted to determine the dimensionality of the test. Exploring the test's unidimensionality, confirmatory factor analyses were performed, followed by higher order factor analyses, and a modified parallel analysis. To explore potential content factors, principal component analyses of the total item set and for each section of the test (SPR and MC) were conducted, as well as a multi-dimensional scaling analysis. / Analyses indicate, to varying degrees, that the test is sufficiently unidimensional for measurement purposes. Results of the second order analysis and the modified parallel analysis offer a qualified yes to whether this test could be considered unidimensional enough to be analyzed using IRT procedures. No format factors were present in the study. The question regarding similarity of content for the two item formats resulted in unexpected process factors for both item formats, however, findings were tentative. / Source: Dissertation Abstracts International, Volume: 54-03, Section: A, page: 0799. / Major Professor: Jacob G. Beard. / Thesis (Ph.D.)--The Florida State University, 1993.
Identifer | oai:union.ndltd.org:fsu.edu/oai:fsu.digital.flvc.org:fsu_76870 |
Contributors | Sheridan, Valerie Barr., Florida State University |
Source Sets | Florida State University |
Language | English |
Detected Language | English |
Type | Text |
Format | 91 p. |
Rights | On campus use only. |
Relation | Dissertation Abstracts International |
Page generated in 0.0018 seconds