It is more common for educational and psychological data to be nonnormal than to be approximately normal. This tendency may lead to bias and error in point estimates of the Pearson correlation coefficient. In a series of Monte Carlo simulations, the Pearson correlation was examined under conditions of normal and nonnormal data, and it was compared…

The aim of this paper is to provide an introduction of new imputation algorithms for estimating missing values from official statistics in larger data sets of data pre-processing, or outliers. The goal is to propose a new algorithm called IRMI (iterative robust model-based imputation). This algorithm is able to deal with all challenges like…

This study compared models of assessment structure for achieving differentiation across the range of examinee attainment in the General Certificate of Secondary Education (GCSE) examination taken by 16-year-olds in England. The focus was on the "adjacent levels" model, where papers are targeted at three specific non-overlapping ranges of…

In many surveys, the data comprise a large number of categorical variables that suffer from item nonresponse. Standard methods for multiple imputation, like log-linear models or sequential regression imputation, can fail to capture complex dependencies and can be difficult to implement effectively in high dimensions. We present a fully Bayesian,…

Person-fit assessment may help the researcher to obtain additional information regarding the answering behavior of persons. Although several researchers examined person fit, there is a lack of research on person-fit assessment for mixed-format tests. In this article, the lz statistic and the ?2 statistic, both of which have been used for tests…

The authors assessed correct model identification rates of Akaike's information criterion (AIC), corrected criterion (AICC), consistent AIC (CAIC), Hannon and Quinn's information criterion (HQIC), and Bayesian information criterion (BIC) for selecting among cross-classified random effects models. Performance of default values for the 5…

A simulation study was conducted to explore the performance of the linear logistic test model (LLTM) when the relationships between items and cognitive components were misspecified. Factors manipulated included percent of misspecification (0%, 1%, 5%, 10%, and 15%), form of misspecification (under-specification, balanced misspecification, and…

Subscores, also known as domain scores, diagnostic scores, or trait scores, can help determine test-takers' relative strengths and weaknesses and appropriately focus remediation. However, subscores often have poor psychometric properties, particularly reliability and distinctiveness (Folske, Gessaroli, & Swanson, 1999; Monaghan, 2006;…

Although the use of simulation to teach the sampling distribution of the mean is meant to provide students with sound conceptual understanding, it may lead them astray. We discuss a misunderstanding that can be introduced or reinforced when students who intuitively understand that "bigger samples are better" conduct a simulation to…

Many studies have been conducted to evaluate the performance of DIF detection methods, when two groups have different ability distributions. Such studies typically have demonstrated factors that are associated with inflation of Type I error rates in DIF detection, such as mean ability differences. However, no study has examined how the direction…

This article is concerned with the question of whether the missing data mechanism routinely referred to as missing completely at random (MCAR) is statistically examinable via a test for lack of distributional differences between groups with observed and missing data, and related consequences. A discussion is initially provided, from a formal logic…

This paper examines the short-run distributional effects of publicly provided education services in Greece using static incidence analysis. Public education is found to be inequality-reducing but the progressivity of the system withers away as we move up to higher educational levels. We employ a framework of both relative and absolute inequality…

When young children attempt to locate the positions of numerals on a number line, the positions are often logarithmically rather than linearly distributed. This finding has been taken as evidence that the children represent numbers on a mental number line that is logarithmically calibrated. This article reports a statistical simulation showing…

It is common to consider Tukey's schematic ("full") boxplot as an informal test for the existence of outliers. While the procedure is useful, it should be used with caution, as at least 30% of samples from a normally-distributed population of any size will be flagged as containing an outlier, while for small samples (N less than 10) even extreme…

Despite the widespread popularity of growth curve analysis, few studies have investigated robust growth curve models. In this article, the "t" distribution is applied to model heavy-tailed data and contaminated normal data with outliers for growth curve analysis. The derived robust growth curve models are estimated through Bayesian…