Program of Events

SM16 was held on Friday, October 28, 2016 at the Memorial Union, MUII. Faculty, postdocs and graduate students from all over campus presented and discussed their quantitative research.

SESSION 1: 9:00 a.m. – 10:30 a.m.

Jingwen Zhang (Communications): “An online experimental approach to study networked social influence in offline behaviors”


Nicole Sparapani (Education): “Profiles of foundational learning skills among first graders obtained using latent class analysis”

Mijke Rhemtulla (Psychology): “Consequences of mistaking the measurement model in structural equation modeling” 

BREAK: 10:30 a.m. – 11:00 a.m. 

SESSION 2: 11:00 a.m. – 12:30 p.m.

Phillipe Rast (Psychology): "Modeling individual differences in within-person variation using a mixed effects location scale model: The effect of physical activity on positive and negative affect" 

Colin Cameron (Economics): “Robust Inference for Regression with Clustered Errors”

Kevin Gee (Education): “Sensitivity analyses for clustered data: An illustration from a large-scale clustered randomized controlled trial in education”


LUNCH: 12:30 p.m. – 1:00 p.m.

SESSION 3: 1:30 p.m. – 3:00 p.m.

C.-Y. Cynthia Lin Lawell (Agricultural and Resource Economics)
: “Investment in corn-ethanol plants in the midwestern United States: A structural econometric model of the dynamic game”

Vasco Yasenov (Economics): “Estimating the lower bound of the impact of Immigrants on natives under the assumption of monotone instrumental variables”

Jaerim Choi (Economics): “Measuring intergenerational income elasticity using weak-instrument robust inference for two-sample instrumental variables regression”

Watch videos of all three sessions

 

PAPER ABSTRACTS

SESSION 1

  • Jingwen Zhang (Communications): “An online experimental approach to study networked social influence in offline behaviors”


Abstract: Online social networks are a highly attractive resource for large scale health initiatives given their capacity to disseminate interventions easily while simultaneously facilitating social influence dynamics. This talk discusses online social networks’ efficacy and mechanisms in increasing physical activity among young adults in three randomized controlled trials.

  • Nicole Sparapani (Education): “Profiles of foundational learning skills among first graders obtained using latent class analysis”

Abstract: Latent Profile Analysis (LPA) is a statistical method for identifying subgroups or profiles of individuals that share characteristics. The current study uses LPA to identify the following four subgroups of students, using foundational learning components, in a sample of first graders: Emergent Hyperactive, Externalizing, Generally Good Students, and Internalizing. These findings provide insight into how researchers and educators might consider students’ profiles of foundational learning components to inform how to support development and learning in the classroom.

  • Mijke Rhemtulla (Psychology): “Consequences of mistaking the measurement model in structural equation modeling”

Abstract: Methodologists often recommend modeling constructs as reflective latent variables to disattenuate them of measurement error. This suggestion has led to widespread, unreflective use of such models, in which scale items are represented as conditionally independent given a latent construct. But the relation between items and constructs can take many forms. I will describe a series of simulations examining the consequences of fitting reflective models to items generated by an alternative model.

SESSION 2

  • Phillipe Rast (Psychology): "Modeling individual differences in within-person variation using a mixed effects location scale model: The effect of physical activity on positive and negative affect"

Abstract: A Bayesian mixed effects location scale model was used to model and explain individual differences in within-person (WP) variability of positive (PA) and negative affect (NA) across two weeks within a measurement burst design. The location scale model allows estimation and prediction of individual differences at both the mean structure level and the residual variance level. That is, the location scale model expands the idea of random effects to the scale part of the model and is ideally suited to address intra-individual variability.

  • Colin Cameron (Economics): “Robust Inference for Regression with Clustered Errors”.

Abstract: This talk considers statistical inference for regression when data are grouped into clusters, with regression model errors independent across clusters but correlated within clusters. Examples include data on individuals with clustering on village or region or other category such as industry, and state-year differences-in-differences studies with clustering on state. In such settings default standard errors can greatly overstate estimator precision. Instead statistical inference OLS should be based on cluster-robust standard errors. The talk outlines the basic method as well as many complications that can arise in practice: cluster-specific fixed effects, few clusters, unbalanced clusters, multi-way clustering, and estimators other than OLS. 

  • Kevin Gee (Education): “Sensitivity analyses for clustered data: An illustration from a large-scale clustered randomized controlled trial in education”


Abstract: Using data from a recently completed cluster RCT of a school-based teacher professional development program, we demonstrate our use of four commonly applied methods for analyzing clustered data. These methods include: (1) hierarchical linear modeling (HLM); (2) feasible generalized least squares (FGLS); (3) generalized estimating equations (GEE); and (4) ordinary least squares (OLS) regression with cluster-robust (Huber–White) standard errors. We compare our findings across each method, showing how inconsistent results – in terms of both effect sizes and statistical significance – emerged across each method and our analytic approach to resolving such inconsistencies 

SESSION 3

  • C.-Y. Cynthia Lin Lawell (Agricultural and Resource Economics): 
“Investment in corn-ethanol plants in the midwestern United States: A structural econometric model of the dynamic game”

Abstract: We model the decision to invest in ethanol plants at the county level using both reduced-form discrete response regression models and a structural model of a dynamic game.  We focus on plants in the Midwestern United States  over the period 1996-2008. We use the estimated structural parameters to simulate counterfactual policy scenarios to disentangle the impacts of state and national policies on the timing and location of investment in the industry. 

  • Vasco Yasenov (Economics): “Estimating the lower bound of the impact of Immigrants on natives under the assumption of monotone instrumental variables.”

Abstract: I apply conservative empirical and theoretical bounding strategies to identify the labor market impacts of immigrants on natives. The estimated bounds on the effects on natives' wages under minimal assumptions rule out elasticities smaller than -0.37 or larger than 0.38. To tighten this interval, I explore mild assumptions motivated by economic theory and instrumental variables which further narrow the lower bound to -0.11. 

  • Jaerim Choi (Economics): “Measuring intergenerational income elasticity using weak-instrument robust inference for two-sample instrumental variables regression”

Abstract: Instrumental variable methods for regression are well-established. More recently methods have been developed for statistical inference when the instruments are weakly correlated with the endogenous regressor so that estimates are no longer asymptotically normally distributed. This paper extends such inference to the case where two separate samples are used to implement instrumental variables estimation. Application is to measuring the extent to which an adult’s income is correlated with their parents’ income.

Watch videos of all three sessions

Read a summary of the 2015 event, including a full list of presenters.