00000namaa22000004a 4500 UP-100084294495502446 Buklod 20230713153937.0 aa grm |||| u ta 230713s2021 ph a rm 00| 0|engdd DEDUC rda eng DMLUC LG 996 2021 E3 M45 Mejia, Ivy P. author Developing a cognitively diagnostic assessment on scientific inquiry under the G-DINA model framework Ivy P. Mejia ; dissertation advisers, Jose Q. Pedrajita, Kevin Carl P. Santos ; dissertation reader, Sheryl Lyn C. Monterola ; Teodora M. Salubayba, Maria Vanessa Lusung-Oyzon, Dennis L. Danipog, members. Quezon City College of Education, University of the Philippines Diliman 2021 xv, 369 leaves illustrations (some color) 28 cm. text rdacontent unmediated rdamedia volume rdacarrier Dissertation (Doctor of Philosophy in Education -- Research and Evaluation) University of the Philippines Diliman February 2021 Includes bibliographic references (177-188 leaves) The study develops a cognitively diagnostic assessment of scientific inquiry under the G-DINA model framework. Specifically, it identifies the underlying structure of the cognitive diagnosis model (CDM) of the use of various forms or evidence to explain phenomena scientifically (UVFEEPS) as obtained from the context of preservice secondary science teachers. Additionally, the objective was to validate whether the theoretical UVFEEPS was supported by the actual model drawn from the result of the diagnostic assessment. To develop a model of cognition of UVFEEPS as the selected construct, an exploratory iterative sequential mixed-method research design was employed. This design has two strands, qualitative and quantitative. For the qualitative strand of the study, the list of attributes of the cognitive diagnosis model on UVFEEPS was obtained from a review of program documents, classroom observations, teacher interviews, verbal reports of students, and experts' views. A cognitive specification per item was also developed which served as the provisional Q-matrix of the assessment. This matrix informed the writing of novel test items designed to diagnose mastery of skills on UVFEEPS. For the quantitative strand of the study, a 40-item cognitively diagnostic assessment on UVFEEPS was developed. The assessment was administered to a sample of 464 preservice science teachers in the four regions. The data from the sample were used to evaluate the fit of the cognitive diagnosis model in the following levels: Q-matrix, test, and items levels. Relative and absolute indices of the cognitive diagnosis models were used for model comparisons at Q-matrix and item level: WALD test for item level, and discrimination index for final item selection. Lastly, the reliability of attributes in the best fitting model was computed using two of the three approaches in classification accuracy: maximum a posteriori (MAP) and expected a posteriori (EAP). The iterative component of the study was done on three occasions. The goal was to refine the Q-matrix on skills on UVFEEPS by comparing the provisional Q-matrix obtained from the qualitative strand and the actual Q-matrix. The findings of the study show that the best fitting cognitive diagnosis model that captures the diagnostic capability of UVFEEPS in the context of preservice science teachers is the mixed-cognitive diagnosis model (a). Findings derived from classification accuracy provide evidence on the reliability of attributes in the mixed-cognitive diagnosis model (a) as the best fitting model. Cognition Testing. Educational tests and measurements Philippines. Teachers Training of Philippines. Cognitive diagnosis model. G-DINA model. Pedrajita, , Jose Q. dissertation adviser. Santos, Kevin Carl P. dissertation adviser. Monterola, Sheryl Lyn C. dissertation reader. Salubayba, Teodora M. member. Oyzon, Maria Vanessa Lusung member. Danipog, Dennis L. member. FI UP UPD DEDUC LG 996 2021 E3 M44 Thesis