Generalized Linear Factor Score Regression: A Comparison of Four Methods

Factor score regression has recently received growing interest as an alternative for structural equation modeling. However, many applications are left without guidance because of the focus on normally distributed outcomes in the literature. We perform a simulation study to examine how a selection of factor scoring methods compare when estimating regression coefficients in generalized linear factor score regression. The current study evaluates the regression method and the correlation-preserving method as well as two sum score methods in ordinary, logistic, and Poisson factor score regression. Our results show that scoring method performance can differ notably across the considered regression models. In addition, the results indicate that the choice of scoring method can substantially influence research conclusions. The regression method generally performs the best in terms of coefficient and standard error bias, accuracy, and empirical Type I error rates. Moreover, the regression method and the correlation-preserving method mostly outperform the sum score methods.     

Perceptions of school and future adjustment to life: A longitudinal study between the ages of 18 and 25

A group of senior high school students, 18 years old, answered a questionnaire about their perceptions of their school situation. Around one-third of the group seemed to like school very much and another third seemed to hate or dislike school very much. This group was followed up until they were 25 years old. Of the original sample 82%, 462 persons, answered a new questionnaire about how they had succeeded in life, what kind of further education they had got, what kind of job they had, how they looked upon their future job and life careers, etc. They were also asked about their memories and feelings about their former school experiences. In the analyses we relate the students' school perceptions at age 18 to their memories of school when 25. A substantial number of the respondents remembered school in the same way as they perceived school when in school. Those who hated school seemed to be more vulnerable when they were adult. We also discuss what could be done to change school so that it does not loose one-third of its students. We introduce the concept of the 'learning centre', the aim of which is to teach the joy of learning and stimulate the students' thirst for knowledge.     

Lower Level Mediation Effect Analysis in Two-Level Studies: A Note on a Multilevel Structural Equation Modeling Approach

This article presents a didactic discussion of a multilevel covariance structure modeling approach to estimation of lowest level mediation effect indexes in two-level studies. The procedure is useful when addressing questions about relations among total and indirect effects between variables of interest while accounting for the hierarchical structure of analyzed data. The discussed method also permits interval estimation and hypothesis tests with respect to related quantities of relevance when evaluating mediated effects with clustered data, and is illustrated on a two-level data set.     

Using LISREL to Fit Nonlinear Latent Curve Models

Latent curve models offer a flexible approach to the study of longitudinal data when the form of change in a response is nonlinear. This article considers such models that are conditionally linear with regard to the random coefficients at the 2nd level. This framework allows fixed parameters to enter a model linearly or nonlinearly, and random coefficients at the 2nd level may only enter linearly. Beginning with LISREL 8.80 for Windows, such models can be fitted, giving users greater flexibility in model specification. An example with LISREL syntax is provided.     

Microfluidic device for generating a stepwise concentration gradient on a microwell slide for cell analysis

Understanding biomolecular gradients and their role in biological processes is essential for fully comprehending the underlying mechanisms of cells in living tissue. Conventional in vitro gradient-generating methods are unpredictable and difficult to characterize, owing to temporal and spatial fluctuations. The field of microfluidics enables complex user-defined gradients to be generated based on a detailed understanding of fluidic behavior at the μm-scale. By using microfluidic gradients created by flow, it is possible to develop rapid and dynamic stepwise concentration gradients. However, cells exposed to stepwise gradients can be perturbed by signals from neighboring cells exposed to another concentration. Hence, there is a need for a device that generates a stepwise gradient at discrete and isolated locations. Here, we present a microfluidic device for generating a stepwise concentration gradient, which utilizes a microwell slide''s pre-defined compartmentalized structure to physically separate different reagent concentrations. The gradient was generated due to flow resistance in the microchannel configuration of the device, which was designed using hydraulic analogy and theoretically verified by computational fluidic dynamics simulations. The device had two reagent channels and two dilutant channels, leading to eight chambers, each containing 4 microwells. A dose-dependency assay was performed using bovine aortic endothelial cells treated with saponin. High reproducibility between experiments was confirmed by evaluating the number of living cells in a live-dead assay. Our device generates a fully mixed fluid profile using a simple microchannel configuration and could be used in various gradient studies, e.g., screening for cytostatics or antibiotics.     

Top-down and bottom-up approaches to improve educational quality: their intended and unintended consequences

Educational Assessment, Evaluation and Accountability -     

The economic microgeography of diversity and specialization externalities – firm-level evidence from Swedish cities

We employ finely geo-coded firm-level panel data to assess the long-standing question whether agglomeration economies derive from specialization (within-industry), diversity (between-industry) or overall density. Rather than treating the city as a single unit, we focus our analysis on how the inner industry structures of cities influence firm-level productivity. Our results illustrate the co-existence of several externalities that differ in their spatial distribution and attenuation within cities. First, we find robust positive effects of neighborhood-level specialization on TFP as well as a small effect of diversity at the same fine spatial level. These effects are highly localized and dissipate beyond the immediate within-city neighborhood level. Second, we also find that firms benefit from the overall density of the wider city. The results emphasize the relevance of “opening up” cities to study the workings of their inner organization and support the idea that location in a within-city industry cluster in a diversified and dense city boosts productivity.     

Robuste Indizes für Projection Pursuit

Die moderne und hoch komplexe Fertigung, z.B. in der Halbleiterindustrie, erfordert für die dabei anfallenden Messdaten der Anlagen- bzw. Prozessparameter multivariate Analysemethoden. Eine dieser m?glichen Analysemethoden ist Projection Pursuit (PP). Dieses Verfahren ist durch geschickte Wahl des so genannten Projektionsindex in der Lage, verschiedenste Datencharakteristika zu detektieren und diese auf anschauliche Weise zu visualisieren. Bei den Projektionsindizes handelt es sich um Funktionen, die eine Projektion auf unterschiedliche Merkmale hin bewerten und dabei jeder Projektion einen Funktionswert zuweisen. Dieser Funktionswert spiegelt die Aussagekraft der Projektion (in Abh?ngigkeit vom verwendeten Index) wider. Die Auswahl bzw. der Aufbau dieser Indizes ist hierbei von entscheidender Bedeutung. Alle hier vorgestellten Indizes zeichnen sich vor allem durch ihren robusten Charakter gegenüber durch Ausrei?er kontaminierte Daten, wie sie in realen Datenszenarien vorkommen, aus. Die durch Anwendung dieser Indizes gewonnenen Einblicke in die komplexe Struktur der Daten erm?glichen es, hochdimensionale Parameterabweichungen und komplexe Zusammenh?nge innerhalb der Daten zu finden.