News and Updates

Categories

How do I know if my structural equation model fits the data well?

January 11, 2019

This is one of the most common questions we receive and, unfortunately, there are no quick answers. However, there are some initial guidelines that can be followed when assessing the fit of an SEM. For most SEMs, the goal of the analysis is to define a model that results in predicted values of the summary statistics (sometimes called “moment structures” consisting of the variances, covariances, and means of the observed variables) that are as close as possible to the values that were observed in the sample. SEMs that result in a closer correspondence between the predicted and the observed summary statistics obtain a better fit compared to models that do not.

(more…)

Read More

Announcing Minimal-Cost Introduction to Multilevel Modeling Workshop for Graduate Students

December 15, 2018

At Curran-Bauer Analytics, we have long been committed to providing broad access to high-quality training opportunities for students in the social, behavioral and health sciences. We are thus very excited to announce a new three-day workshop, Introduction to Multilevel Modeling for Graduate Students, to be held in Chapel Hill, NC on May 29-31, 2019, at steeply reduced cost exclusively for students actively enrolled in a graduate or professional degree granting academic program. Students pay only a $100 processing fee to attend. Enrollment is limited to 35 individuals and participants will be selected by random lottery for admission if applications exceed this number. Pre-register by March 15th to enter the admission lottery. Click the link above for details and to pre-register.

Read More

2019 Summer Workshops

August 28, 2018

We’re pleased to announce our summer workshop schedule for 2019:

New this year: We have expanded Network Analysis from three to five days to allow more in-depth coverage, we now provide live demonstrations in R for Structural Equation Modeling and Longitudinal Structural Equation Modeling, and we have expanded our R demonstrations for Latent Class/Cluster Analysis and Mixture Modeling.

See our Training page for a general description of our teaching philosophy, links to course reviews and sample course notes.

Read More

The Importance of Studying Individual Trajectories, Even for Countries

May 31, 2018

The analysis of longitudinal data has quickly gained in importance across a variety of fields because it allows for the examination of questions about change over time. This is why all of our current workshops (Network Analysis, Latent Class/Mixture Modeling, Multilevel Modeling, Structural Equation Modeling, and Longitudinal Structural Equation Modeling) address the analysis of longitudinal data to some degree. A common theme underlying nearly any analysis of repeated measures data is the importance of modeling between-unit differences in within-unit change. We use the term “unit” instead of “person” because these models can be applied to repeated measures that have been drawn from any unit of observations, whether that is an individual person or region of the brain or even country. This latter application is clearly demonstrated in a recent article in the New York Times examining the relation between health expenditures and health outcomes over time and across country. The article presents several highly impactful graphics that clearly show that the United States was characterized by a trajectory of life expectancy that was similar to a number of comparable countries, but only up to about 1980. After that time the US has systematically fallen behind comparison countries, even though health care spending has increased during this same period. Of course the identification of the specific causal influences leading to these changes is complex, but only through the analysis of the repeated measures data on both health care spending and life expectancy could these trends be identified. Developing an understanding of these trajectories in turn allows for the generation of hypotheses about the causal mechanisms and the implementation of changes in policy that might lead to improvements in life expectancy in the future. It is thus important to consider the analysis of individual trajectories, regardless of what “individual” means in your own research application.

Read More

Best Methods for Handling Missing Data in Intensive Longitudinal Designs

April 13, 2018

In nearly every discipline within the behavioral, health, and educational sciences, longitudinal data have become requisite for establishing temporal precedence and distinguishing inter-individual differences in intra-individual change. Whereas traditional longitudinal designs often obtained repeated assessments at monthly or even yearly intervals, recent advances in mobile technology have allowed for the collection of multiple assessments throughout a single day. These so-called intensive longitudinal designs (ILDs) are becoming increasing prevalent in many empirical studies of human development and behavior. However, as with any advancement in design and assessment, a multitude of complexities arise when fitting statistical models to large numbers of repeated assessments often taken on smaller numbers of individuals. For example, it is not uncommon to use an ILD to obtain six daily assessments over a 14 day period on 75 individuals. Key among a variety of complexities that must be addressed is that of missing data. Standard existing methods for handling missing data are not always well suited when applied to large numbers of repeated assessments and guidance for practitioners is sparse. A recent paper in the journal Structural Equation Modeling by Linying Ji and colleagues addresses this very issue. The paper is motivated by an actual ILD in which a large number of assessments are obtained from parents in the assessment of emotional states and behaviors arising from conflicts between parents and children. In the original application, missing data are pervasive yet no well developed methods were available to address this issue. The authors then describe several modern methods for the analysis of missing data in ILDs, they conduct a computer simulation to evaluate the methods under known conditions, and they re-analyze the empirical data to demonstrate the new techniques. They conclude with recommendations for handling missing data in ILDs and provide R code to help in this endeavor.

Read More