Christopher Adah Ocheni1 (https://orcid.org/0000-0001-6007-2362), John Joseph Agah2 (https://orcid.org/0000-0001-9076-0887), Anayo David Nnaji3 (https://orcid.org/0009-0002-1635-7525), Basil Chinecher
COMPARING POLYTOMOUS ITEM RESPONSE THEORY MODELS FOR COGNITIVE LOAD: EVIDENCE FROM PARTIAL CREDIT AND GENERALIZED PARTIAL CREDIT MODELS
https://doi.org/10.83151/g0ry-ck77
The focus of the study was to empirically compare the sensitivities of the partial credit and generalized partial credit polytomous item response theory models for cognitive load. Methodologically, a psychometric analysis research design was adopted in this study. Three research questions guided the study. This study used secondary data collected from a sample of 307 (186 males and 121 females) Nigerian secondary school students through an online survey. The data were collected using the Cognitive Load Rating Scale (CLRS). The data was analysed using descriptive statistics such as mean, standard deviation, kurtosis, and skewness, while marginal and empirical reliability indices were used to establish the internal structure and consistency of the items. The Partial Credit Model (PCM) and Generalized Partial Credit Model (GPCM) were used for the analyses of data. The findings of the study showed, among others, that both models provided acceptable fit estimates; however, the GPCM was a better fit to the data. It was concluded that the CLRS shows sound psychometric properties, with ordered categories, acceptable reliability, and adequate fits for most items. Based on the findings, it was recommended that educational researchers and practitioners should rely on the GPCM for assessing the model fit and other psychometric properties of the CLRS.
Keywords: Cognitive Load, Rating Scale, Partial Credit Model, Generalized Partial Credit, Model Fit, Polytomous IRT, Rasch model
Christopher Adah Ocheni1 (https://orcid.org/0000-0001-6007-2362), John Joseph Agah2 (https://orcid.org/0000-0001-9076-0887), Anayo David Nnaji3 (https://orcid.org/0009-0002-1635-7525), Basil Chinecher
COMPARING POLYTOMOUS ITEM RESPONSE THEORY MODELS FOR COGNITIVE LOAD: EVIDENCE FROM PARTIAL CREDIT AND GENERALIZED PARTIAL CREDIT MODELS https://doi.org/10.83151/g0ry-ck77
The focus of the study was to empirically compare the sensitivities of the partial credit and generalized partial credit polytomous item response theory models for cognitive load. Methodologically, a psychometric analysis research design was adopted in this study. Three research questions guided the study. This study used secondary data collected from a sample of 307 (186 males and 121 females) Nigerian secondary school students through an online survey. The data were collected using the Cognitive Load Rating Scale (CLRS). The data was analysed using descriptive statistics such as mean, standard deviation, kurtosis, and skewness, while marginal and empirical reliability indices were used to establish the internal structure and consistency of the items. The Partial Credit Model (PCM) and Generalized Partial Credit Model (GPCM) were used for the analyses of data. The findings of the study showed, among others, that both models provided acceptable fit estimates; however, the GPCM was a better fit to the data. It was concluded that the CLRS shows sound psychometric properties, with ordered categories, acceptable reliability, and adequate fits for most items. Based on the findings, it was recommended that educational researchers and practitioners should rely on the GPCM for assessing the model fit and other psychometric properties of the CLRS. Keywords: Cognitive Load, Rating Scale, Partial Credit Model, Generalized Partial Credit, Model Fit, Polytomous IRT, Rasch model