1 January 2021 in Open Education Studies
In educational measurement, various methods have been proposed to infer student proficiency from the ratings of multiple items (e.g., essays) by multiple raters. However, suitable models quickly become numerically demanding or even unfeasible as separate latent variables are needed to account for local dependencies between the ratings of the same response. Therefore, in the present paper we derive a flexible approach based on Thurstone’s law of categorical judgment. The advantage of this approach is that it can be fit using weighted least squares estimation which is computationally less demanding as compared to most of the previous approaches in the case of an increasing number of latent variables. In addition, the new approach can be applied using existing latent variable modeling software. We illustrate the model on a real dataset from the Trends in International Mathematics and Science Study (TIMMSS) comprising ratings of 10 items by 4 raters for 150 subjects. In addition, we compare the new model to existing models including the facet model, the hierarchical rater model, and the hierarchical rater latent class model.
10 May 2017 in Education and Science
The purpose of this research is to determine whether student oriented teaching, experience oriented teaching, teacher support and the class size predict the usage of formative assessment in mathematics. This study is designed as a predictive research that falls in the correlational survey model, one of the general survey models. The sample of the study consists of PISA 2012 Turkey data (4848 students). The data were obtained from the students and school questionnaires used within the scope of PISA 2012. The data were analyzed by using ordinal logistic regression analysis. The variables of the study consist of two group (predicted and predictor) variables. The predicted variable is formative assessment use and the predictor variables are student oriented teaching, experience oriented teaching, teacher support and the class size, respectively. This study has proved that while the variables “student oriented teaching” and “teacher support” are significant predictors of how much formative assessment is used, the variables “experience oriented teaching” and “class size” are not significant predictors of how much formative assessment is used. Nevertheless, the logistic regression model designed proved to be meaningful as a whole. The rate of the classification accuracy of the model is 58%.
25 December 2022 in Eğitimde ve Psikolojide Ölçme ve Değerlendirme Dergisi
Currently, research on computerized adaptive testing (CAT) focuses mainly on dichotomous items and cognitive traits (achievement, aptitude, etc.). However, polytomous IRT-based CAT is a promising research area for measuring psychological traits that has attracted much attention. The main purpose of this study is to test the practicality of the polytomous IRT-based CAT and its equivalence with the paper-pencil version. Data were collected from 1449 high school students (45% female) via the paper-pencil version. The data were used for IRT parameter estimates and CAT simulation studies. For the equivalence study, the research group consisted of 81 students (47% female) who participated in both the paper-pencil and live CAT applications. The paper-pencil version of the vocational interest inventory consists of 17 factors and 164 items. When the EAP estimation method and setting SE < .50 as the termination criterion, better performance was obtained compared with other CAT designs. The Item selection did not help to reduce test duration or increase measurement accuracy. As a result, it was found that an area of interest can be assessed with four items. The results of the live CAT application showed that the estimates of CAT were strongly positively correlated with its paper-pencil version. In addition, the live CAT application increased applicability compared to the fixed-length test version by reducing test length by 50% and time by 77%. This study shows that the polytomous IRT-based CAT is applicable and efficient for measuring psychological traits.
16 March 2020 in Öneri Dergisi
Kovaryans analizi özünde, istenmeyen değişkenlerin bağımlı değişkene yaptığı etkilerin doğrusal regresyonla dengelendiği varyans analizidir. Bu analizde önce doğrusal regresyon yapılarak istenmeyen değişkenlerin gruplara etkileri hesaplanmaktadır. Regresyon katsayıları hesaplandıktan sonra da, bu katsayılar kullanılarak bağımlı değişkenden istenmeyen değişkenlerin (kontrol değişkenlerinin) hesaplanan etkileri çıkarılmakta ve bu şekilde dengelenmiş olan verilere varyans analizi uygulanarak daha sağlıklı sonuçlar elde edilmektedir.