Jacob Bishop

Testing the flipped classroom with model-eliciting activities and video lectures in a mid-level undergraduate engineering course

This paper outlines an ongoing study of the flipped classroom with second and third-year undergraduate engineering students in a numerical methods course. The flipped classroom is a new pedagogical method, which employs asynchronous video lectures and practice problems as homework, and active, group-based problem-solving activities in the classroom. It represents the combination of learning theories once thought to be incompatible-active, problem-based learning activities founded upon constructivist ideals and instructional lectures derived from direct instruction methods founded upon behaviorist principles. Using a controlled quasi-experimental research design, we conduct a study with a full 15-week numerical methods course at Utah State University during the spring semester of 2013. Students in the experimental section completed model-eliciting activities inside the classroom and video lectures and homework outside the classroom. Students in the control section completed homework outside the classroom and group lectures inside the classroom. The two groups will be compared using scores from homework, examinations, and a sixteen-question numerical methods conceptual pre- post- test pair. The three main features that distinguish this study from previous research are: 1) This is a controlled study; 2) This study examines student performance on objective measures; 3) This study uses model-eliciting activities in the experimental classroom.

Modeling latent growth with multiple indicators: a comparison of three approaches.

Latent growth curve models (LGCMs) are widely used methods for analyzing change in psychology and the social sciences. To date, most applications use first-order (single-indicator) LGCMs. These models have several limitations that can be overcome with multiple-indicator LGCMs. Currently, almost all multiple-indicator applications use the so-called second-order growth model (SGM; McArdle, 1988). In this article, we review the SGM and discuss 2 alternative, but less well-known, multiple-indicator LGCMs that overcome some of the limitations of the SGM: the generalized second-order growth model (GSGM) and the indicator-specific growth model (ISGM). In contrast to the SGM, the GSGM does not involve a proportionality constraint on the ratio of general to specific variance. The ISGM allows researchers to model indicator-specific growth. Both of these alternative models allow testing measurement invariance across time for state-variability components. We also present an empirical application regarding changes in self-reported levels of anxiety and discuss implications of the differences between the 3 models for applied research.

Analyzing person, situation and person × situation interaction effects: Latent state-trait models for the combination of random and fixed situations.

Latent state-trait (LST) models (Steyer, Ferring, & Schmitt, 1992) allow separating person-specific (trait) effects from (1) effects of the situation and person × situation interactions, and (2) random measurement error in purely observational studies. Typical LST applications use measurement designs in which all situations are sampled randomly and do not have to be known for any individual. Limitations of conventional LST models for only random situations are that traits are implicitly assumed to generalize perfectly across situations, and that main effects of situations are inseparable from person × situation interaction effects because both are measured by the same latent variable. In this article, we show how these limitations can be overcome by using measurement designs in which two or more random situations are nested within two or more fixed situations that are known for each individual. We present extended LST models for the combination of random and fixed situations (LST-RF approach) and show that the extensions allow (1) examining the extent to which traits are situation-specific and (2) isolating person × situation interactions from situation main effects. We demonstrate that the LST-RF approach can be applied with both homogenous and heterogeneous indicators in either the single- or multilevel structural equation modeling frameworks. Advantages and limitations of the new models as well as their relation to other approaches for studying person × situation interactions are discussed.

Analyzing latent state-trait and multiple-indicator latent growth curve models as multilevel structural equation models

Latent state-trait (LST) and latent growth curve (LGC) models are frequently used in the analysis of longitudinal data. Although it is well-known that standard single-indicator LGC models can be analyzed within either the structural equation modeling (SEM) or multilevel (ML; hierarchical linear modeling) frameworks, few researchers realize that LST and multivariate LGC models, which use multiple indicators at each time point, can also be specified as ML models. In the present paper, we demonstrate that using the ML-SEM rather than the SL-SEM framework to estimate the parameters of these models can be practical when the study involves (1) a large number of time points, (2) individually-varying times of observation, (3) unequally spaced time intervals, and/or (4) incomplete data. Despite the practical advantages of the ML-SEM approach under these circumstances, there are also some limitations that researchers should consider. We present an application to an ecological momentary assessment study (N = 158 youths with an average of 23.49 observations of positive mood per person) using the software Mplus (Muthén and Muthén, 1998–2012) and discuss advantages and disadvantages of using the ML-SEM approach to estimate the parameters of LST and multiple-indicator LGC models.

Collapsing factors in multitrait-multimethod models: examining consequences of a mismatch between measurement design and model

Models of confirmatory factor analysis (CFA) are frequently applied to examine the convergent validity of scores obtained from multiple raters or methods in so-called multitrait-multimethod (MTMM) investigations. Many applications of CFA-MTMM and similarly structured models result in solutions in which at least one method (or specific) factor shows non-significant loading or variance estimates. Eid et al. (2008) distinguished between MTMM measurement designs with interchangeable (randomly selected) vs. structurally different (fixed) methods and showed that each type of measurement design implies specific CFA-MTMM measurement models. In the current study, we hypothesized that some of the problems that are commonly seen in applications of CFA-MTMM models may be due to a mismatch between the underlying measurement design and fitted models. Using simulations, we found that models with M method factors (where M is the total number of methods) and unconstrained loadings led to a higher proportion of solutions in which at least one method factor became empirically unstable when these models were fit to data generated from structurally different methods. The simulations also revealed that commonly used model goodness-of-fit criteria frequently failed to identify incorrectly specified CFA-MTMM models. We discuss implications of these findings for other complex CFA models in which similar issues occur, including nested (bifactor) and latent state-trait models.