New Confidence Intervals and Bias Comparisons Show That Maximum Likelihood Can Beat Multiple Imputation in Small Samples

When analyzing incomplete data, is it better to use multiple imputation (MI) or full information maximum likelihood (ML)? In large samples ML is clearly better, but in small samples ML’s usefulness has been limited because ML commonly uses normal test statistics and confidence intervals that require large samples. We propose small-sample t-based ML confidence intervals that have good coverage and are shorter than t-based confidence intervals under MI. We also show that ML point estimates are less biased and more efficient than MI point estimates in small samples of bivariate normal data. With our new confidence intervals, ML should be preferred over MI, even in small samples, whenever both options are available.     

“Muslims are not Terrorists”: Islamic State Coverage,Journalistic Differentiation Between Terrorism and Islam,Fear Reactions,and Attitudes Toward Muslims

Previous research shows that news about Islamist terrorism can seriously affect citizens’ fear reactions and influence non-Muslims’ out-group perceptions of Muslims. We argue that news coverage that explicitly links Islam to terrorism or terrorists of the Islamic State (IS) may trigger fears in non-Muslim individuals. In contrast, news differentiation (i.e., explicitly distinguishing between Muslims and Muslim terrorists) may dampen particular fear reactions in citizens. To test the specific effects of news differentiation, a controlled laboratory experiment was conducted. Results showed that undifferentiated news about IS terrorism increased participants’ fear of terrorism and resulted in hostile perceptions toward Muslims in general. However, fear of terrorism only enhanced hostile attitudes toward Muslims for individuals with negative and moderately positive prior experiences with Muslims. For those with very positive experiences, no such relationship was found. Implications of these findings for journalism practice and intergroup relations in multicultural societies are discussed.     

Theoretical framework and empirical evidence of students' cognitive processes in three dimensions of content,complexity, and time

In the literature, learners' cognitive development is mainly discussed with respect to changes in learners' content‐dependent knowledge (conceptual change or growth). Additional dimensions of time and complexity may also be taken into account to describe cognitive processes in at least three dimensions. We discuss these three dimensions of content, complexity, and time theoretically and show empirically what results can be found with such a framework. The primary database consists of 27 students who worked in groups of three on physics tasks during three sessions in an experimental setting. All activities were videotaped and analyzed in great detail. Results indicate that students' meanings always refer to a narrow area of content, are developed bottom‐up with respect to complexity, and show time‐dependent dynamics in which maxima of 30 seconds and 5 minutes are important time scales. Comparison with outcomes of additional experimental settings on the same topic with more and less advanced students indicate the results to have a greater validity. © 2003 Wiley Periodicals, Inc. J Res Sci Teach 40: 616–648, 2003     

Constructivism in Practice: an Exploratory Study of Teaching Patterns and Student Motivation in Physics Classrooms in Finland,Germany and Switzerland

For the last three decades, moderate constructivism has become an increasingly prominent perspective in science education. Researchers have defined characteristics of constructivist-oriented science classrooms, but the implementation of such science teaching in daily classroom practice seems difficult. Against this background, we conducted a sub-study within the tri-national research project Quality of Instruction in Physics (QuIP) analysing 60 videotaped physics classes involving a large sample of students (N?=?1192) from Finland, Germany and Switzerland in order to investigate the kinds of constructivist components and teaching patterns that can be found in regular classrooms without any intervention. We applied a newly developed coding scheme to capture constructivist facets of science teaching and conducted principal component and cluster analyses to explore which components and patterns were most prominent in the classes observed. Two underlying components were found, resulting in two scales—Structured Knowledge Acquisition and Fostering Autonomy—which describe key aspects of constructivist teaching. Only the first scale was rather well established in the lessons investigated. Classes were clustered based on these scales. The analysis of the different clusters suggested that teaching physics in a structured way combined with fostering students’ autonomy contributes to students’ motivation. However, our regression models indicated that content knowledge is a more important predictor for students’ motivation, and there was no homogeneous pattern for all gender- and country-specific subgroups investigated. The results are discussed in light of recent discussions on the feasibility of constructivism in practice.     

The Chain and Post-Stratification Methods for Observed-Score Equating: Their Relationship to Population Invariance

The Non-Equivalent-groups Anchor Test (NEAT) design has been in wide use since at least the early 1940s. It involves two populations of test takers, P and Q, and makes use of an anchor test to link them. Two linking methods used for NEAT designs are those (a) based on chain equating and (b) that use the anchor test to post-stratify the distributions of the two operational test scores to a common population (i.e., Tucker equating and frequency estimation). We show that, under different sets of assumptions, both methods are observed score equating methods and we give conditions under which the methods give identical results. In addition, we develop analogues of the Dorans and Holland (2000) RMSD measures of population invariance of equating methods for the NEAT design for both chain and post-stratification equating methods.     

Using Networks to Visualize and Analyze Process Data for Educational Assessment

New technology enables interactive and adaptive scenario‐based tasks (SBTs) to be adopted in educational measurement. At the same time, it is a challenging problem to build appropriate psychometric models to analyze data collected from these tasks, due to the complexity of the data. This study focuses on process data collected from SBTs. We explore the potential of using concepts and methods from social network analysis to represent and analyze process data. Empirical data were collected from the assessment of Technology and Engineering Literacy, conducted as part of the National Assessment of Educational Progress. For the activity sequences in the process data, we created a transition network using weighted directed networks, with nodes representing actions and directed links connecting two actions only if the first action is followed by the second action in the sequence. This study shows how visualization of the transition networks represents process data and provides insights for item design. This study also explores how network measures are related to existing scoring rubrics and how detailed network measures can be used to make intergroup comparisons.     

How to improve argumentation comprehension in university students: experimental test of a training approach

Instructional Science - The ability to comprehend informal arguments is essential for scientific literacy but students often lack structural knowledge about these arguments, especially when the...     

Exclusion Expected? Cardiac Slowing Upon Peer Exclusion Links Preschool Parent Representations to School-Age Peer Relationships

Attachment theory proposes that children’s representations of interactions with caregivers guide information-processing about others, bridging interpersonal domains. In a longitudinal study (N = 165), preschoolers (M_age = 5.19 years) completed the MacArthur Story Stem Battery to assess parent representations. At school-age (M_age = 8.42 years), children played a virtual ballgame with peers who eventually excluded them to track event-related cardiac slowing, a physiological correlate of rejection, especially when unexpected. At both ages, parents and teachers reported on peer and emotional problems. During exclusion versus inclusion-related events, cardiac slowing was associated with greater positive parent representations and fewer emerging peer problems. Cardiac slowing served as a mediator between positive parent representations and peer problems, supporting a potential psychophysiological mechanism underlying the generalization of attachment-related representations to peer relationships.