In confirmatory factor analysis (CFA), which is used quite often for scale development and adaptation studies, the selected estimation method, affects the results obtained from the data. Because of the selected estimation method, the model parameters and their standard errors, and the model data fit values may alter the results substantially. So that, the purpose of this research is to compare the performance of different estimation methods for CFA. Maximum likelihood (ML), unweighted least squares (ULS) and diagonally weighted least squares (DWLS) are used in this research as estimation methods. These methods are applied in data sets and regression coefficients and their standard errors, t values, fit indexes and iteration numbers obtained from these estimation methods are examined. As a result, ULS method can converge with the minimum number iterations and it seems to be the more accurate method for estimating the parameters.

Stress is defined as a person's interaction with their environment that is thought to threaten or affect an individual's potential, resources, and well‐being. The most popular instrument to assess perceived stress is the Perceived Stress Scale (PSS). Therefore, making a systematic review of studies testing the internal structure of PSS and conducting a Meta‐Analytic Confirmatory Factor Analysis (MACFA) on the database created with the information obtained from these studies are the aims of this research. A total of 76 samples from 57 unique studies were included in this database using various inclusion criteria (total N for PSS‐14 = 28,632, for PSS‐10 = 46,053). The correlated two‐factor model for PSS was confirmed by MACFA performed on the pooled correlation matrix generated by the random effects meta‐analysis. Findings of dimensionality analyses, factor loadings, omega values, and measurement invariance showed that the model that best explained the factor structure of PSS was the correlated two‐factor model. The strict measurement invariance of the PSS was achieved across age and clinical status, and the internal consistency was high according to the omega values. Several recommendations moving forward are discussed.

This study attempted to identify the information and communication technology items that affected students' mathematics and science literacy scores by making use of the 2015 PISA data, The presence of numerous items related to ICT in the PISA and the administration of these items to large groups of people provides researchers with a large data source. However, researchers experience challenges in revealing the significant and beneficial data among the entire data set. So one of the most commonly used data mining method is the Chi-squared Automatic Interaction Detection method (CHAID), which is the decision tree method. As a result of the CHAID analysis, conducted to reveal the ICT items related to mathematics literacy scores, it was revealed that there was a significant relationship between mathematics literacy scores and the eight variables. For science literacy, there was a ten significant relationship variables. There is a relationship between high science and mathematics literacy scores and using digital devices at an early age as well as feeling comfortable with using digital devices at home. As an outcome of the CHAID algorithm, the realization of a significant reduction was achieved in the dimensionality of both models. The selected variables can be used for future research and development of new, parametric models. In the resulting model, apart from the reduction of the number of predictors, the reduction of their categories was also achieved.

The purpose of this research is to first estimate the item and ability parameters and the standard error values related to those parameters obtained from Unidimensional Item Response Theory (UIRT), bifactor (BIF) and Testlet Response Theory models (TRT) in the tests including testlets, when the number of testlets, number of independent items, and sample size change, and then to compare the obtained results. Mathematic test in PISA 2012 was employed as the data collection tool, and 36 items were used to constitute six different data sets containing different numbers of testlets and independent items. Subsequently, from these constituted data sets, three different sample sizes of 250, 500 and 1000 persons were selected randomly. When the findings of the research were examined, it was determined that, generally the lowest mean error values were those obtained from UIRT, and TRT yielded a mean of error estimation lower than that of BIF. It was found that, under all conditions, models which take into consideration the local dependency have provided a better model-data compatibility than UIRT, generally there is no meaningful difference between BIF and TRT, and both models can be used for those data sets. It can be said that when there is a meaningful difference between those two models, generally BIF yields a better result. In addition, it has been determined that, in each sample size and data set, item and ability parameters and correlations of errors of the parameters are generally high.

This study is to examine the meta-analysis results acquired from Cronbach alpha reliability coefficient being used in Hacettepe University Journal of Education. Within this context, 1222 items taking place in 43 issues which were published in Hacettepe University Journal of Education between 1986 and 2012 were examined and 354 measurement tools in total were discussed according to their inclusion criteria. In this study, r index was used in calculating influence quantity in correlational data for combining data while random influence model of Fisher z method was used in correlational data for combining influence quantities. It was benefited from SPSS 20 and MetaWin 2.0 packet programs for the analysis of data. It was found in this study that effect size mean is quite strong according to various moderator varieties of Cronbach alpha reliability coefficient. As a result of this study, it was determined that effect size mean of scale reliability prepared for measuring affective structure was greater than the scales prepared for measuring cognitive structures; effect size mean of reliability coefficient in adaptation studies was greater than the studies of applying and developing prepared scale. In addition to these results, it was observed that as the education levels of individuals within the sample and the number of items in scale increase, effect size mean of alpha reliability coefficient increases, too. On the other hand, it was determined that sample size and answer category number of option items did not have a direct influence on alpha coefficient. Afterwards, independent samples t test, one way ANOVA and Kruskal Wallis Test were conducted in order to find out whether the mean of Cronbach alpha acquired in line with the determined moderator variables varied according to groups. According to these analysis results, it was determined that structure of data collection tool, content of the sample, item and number of option had significant differences on Cronbach alpha coefficient. In addition to the variables within this study, different variables which are thought to affect reliability can be handled and the effects of these variables on reliability coefficient can be examined. Keywords: Cronbach alpha, meta-analysis, reliability ÖZ: Bu çalışmada, Hacettepe Üniversitesi Eğitim Fakültesi Dergisi'nde yayınlanan çalışmalarda yer alan Cronbach alfa güvenirlik katsayılarından elde edilen meta analiz sonuçlarının incelenmesi amaçlanmıştır. Bu amaçla 1986-2012 yılları arasında Hacettepe Üniversitesi Eğitim Fakültesi Dergisi'nde yayınlanan 43 sayıda yer alan toplam 1222 makale incelenmiş ve dahil edilme ölçütlerine göre toplam 354 ölçme aracı ele alınmıştır. Bu araştırmada verilerin birleştirilmesi için korelasyonel verilerde etki büyüklüğü hesaplamalarındaki r indeksi; korelasyonel verilerde etki büyüklüklerinin birleştirilmesi için de Fisher z yönteminin tesadüfi etki modeli kullanılmıştır. Verilerin analizi için SPSS 20 ve MetaWin 2.0 paket programından yararlanılmıştır. Çalışmada Cronbach alfa güvenirlik ...