Selected Publications

(2023). Using robust standard errors for the analysis of binary outcomes with a small number of clusters. JREE.

PDF Code Dataset

(2023). Accounting for heteroskedasticity resulting from between-group differences in multilevel models. MBR.

Code

(2023). Alternatives to logistic regression models when analyzing cluster randomized trials with binary outcomes. Prev Sci.

PDF

(2023). Seasonality of school climate. School Psychology Review.

(2023). The impact of restorative practices on the use of out-of-school suspensions: Results from a cluster randomized controlled trial. Prevention Science.

PDF

(2023). Practical Multilevel Modeling Using R. Sage Publishing.

Preprint PDF

(2023). Advanced categorical data analysis in prevention science. Prevention Science.

PDF

(2022). Evaluation of the Whole School Restorative Practices Project: One-year impact on discipline incidents. JSP.

PDF

(2022). Using cluster robust standard errors when analyzing cluster randomized trials with few clusters. BRM.

Preprint PDF Code Dataset

(2022). Alternatives to logistic regression models in experimental studies. JXE, 90(1).

PDF Code Dataset

(2022). Analyzing cross-sectionally clustered data using generalized estimating equations. JEBS.

Preprint PDF Code Dataset

Recent Publications

See CV for complete list

More Publications

(2023). Steadfast restorative and supportive approaches to student behavior during the COVID-19 pandemic. JEPC.

(2023). Usability and social consequences of the early identification system as a universal screener for social, emotional, and behavioral risks. In SP.

PDF

(2023). In-School and Out-of-School Suspension: Behavioral and Psychological Outcomes in a Predominately Black Sample of Middle School Students. School Psychology Review, doi: https://doi.org/10.1080/2372966X.2020.1856628.

PDF

Recent & Upcoming Talks

More Talks

Multilevel Modeling with Large-Scale International Databases Using HLM (Philadelphia)
Apr 10, 2024 9:00 AM
American Educational Research Association
Using cluster robust standard errors to analyze nested data with a few clusters (Korea)
Oct 12, 2023 2:00 PM
23rd International Conference on Education Research
PDF
Using Cluster Robust Standard Errors to Analyze Cross-Classified Data with a Small Number of Clusters
Sep 29, 2023 2:00 PM
Society for Research on Educational Effectiveness
PDF
Multilevel Modeling with Large-Scale International Databases Using HLM (Chicago)
Apr 12, 2023 9:00 AM
American Educational Research Association
Alternatives to logistic regression models in the analysis of cluster randomized trials with binary outcomes
Jun 4, 2021 10:00 AM
Society for Prevention Research Annual Meeting
PDF Slides
Multilevel Modeling with Large-Scale International Databases Using HLM
Apr 15, 2021 1:00 PM
American Educational Research Association
Multilevel Modeling with Large-Scale International Databases Using HLM
Apr 16, 2020 9:00 AM
American Educational Research Association
Alternatives to logistic regression models in experimental studies
Feb 21, 2020 10:00 AM
PDF Slides
Instrumental Variables for Educational Researchers
Sep 20, 2019 9:00 AM
PDF Slides
Summer of R Workshop
Aug 12, 2019 9:00 AM
Summer of R Workshop
Slides

Recent Posts

More Posts

Last updated on 3 min read jobs4

List of Research, Statistics, and Evaluation job postings (that I’ve seen) as of 2023-11-28. Postings for (2023-2024): As of 2023.11 Tenure-Track Assistant Professor position in Education Statistics and Data Science, The University of Delaware Assistant Professor of Quantitative Research Methods, Tenure-Track, Chapman University, Orange, CA Assistant/Associate Professor - Educational Research, Measurement, and Analysis, Auburn University, Auburn, AL Assistant/Associate Professor in Quantitative Research Methodology, Oklahoma State University, Stillwater, Oklahoma

CONTINUE READING

Last updated on 4 min read regression

Random notes. Regression based techniques often involve finding a maximum (e.g., the maximum likelihood) or a minimum (e.g., least squares or mean square error) value. Gradient descent is an iterative optimization algorithm used to find the minimum of a function (or gradient ascent to find the maximum). The algorithm for solving for \(\theta_j\) looks like: \[\theta_j = \theta_j - \alpha\frac{\partial{}}{\partial{\theta_j} }J(\theta)\] \(\alpha\) is the learning rate (smaller step size takes more iterations)

CONTINUE READING

Last updated on 8 min read residuals

(MLM notes). Residuals are often used for model diagnostics or for spotting outliers in the data. For single-level models, these are merely the observed - predicted data (i.e., \(y_i - \hat{y}_i\)). However, for multilevel models, these are a bit more complicated to compute. Since we have two error terms (in this example), we will have two sets of residuals. For example, a multilevel model with a single predictor at level one can be written as:

CONTINUE READING

Last updated on 2 min read residuals

Random stuff II: Plotting residuals I was poking around my old teaching files and I found an old file and I wasn’t sure what it was: dat <- read.table("https://raw.githubusercontent.com/flh3/pubdata/main/Stefanski_2007/mizzo_1_data_yx1x5.txt") head(dat) ## V1 V2 V3 V4 V5 V6 ## 1 -0.22391 0.0054599 0.380310 0.0135140 0.209240 0.1467100 ## 2 0.84413 0.1073700 -0.026533 0.0458640 0.012987 -0.0271900 ## 3 1.06240 0.0911160 0.181260 0.0501710 -0.188670 -0.0120820 ## 4 -1.04170 0.4404900 0.245960 0.0054154 -0.212920 0.1015200 ## 5 0.

CONTINUE READING

Last updated on 2 min read simulation

Random stuff: Rademacher distribution A month or two ago, I saw this comic on xkcd: This just shows how folks might use different ways to say the same thing (something simple can appear complex and vice versa). And then, in that same week, I was reading an article that mentioned drawing a random variable from a Rademacher distribution– which was a distribution I had not heard of (w/c has a M = 0, SD = 1) and somehow the name alone made it sound complicated.

CONTINUE READING

Projects

All School climate Youth violence project Grants

Ongoing Grant Funded Work

Various ongoing grants funded by the National Institute of Justice, Department of Education (i3), and Institute of Education Sciences.

Missouri Prevention Science Institute

The Missouri Prevention Science Institute (MPSI) brings community members and researchers together to help schools and families apply techniques that promote social and academic success. Through community outreach, the institute’s staff provides parent training and teacher consultation services.

Youth Violence Project @ UVA

Our team of faculty and graduate students conducts research on effective methods and policies for youth violence prevention and school safety.

Teaching

I teach the following courses at the University of Missouri:

Multilevel Modeling

The goal of the course is to provide students with the necessary skills needed to review/critique, analyze, interpret, and write-up studies involving nested (clustered) data using multilevel modeling (MLM). Clustered data (e.g., students within schools, patients within clinics) occur quite naturally in the social sciences and being able to understand and conduct their own analyses using nested data is an important skill. Alternatives are discussed as well.

Applied Multivariate Statistics

This course is designed to provide students with both a theoretical and applied understanding of useful multivariate statistical procedures (e.g., factor analysis, principal components analysis, discriminant function analysis, cluster analysis, MANOVA) in education sciences.

Program (Impact) Evaluation

Evaluating the quantifiable impact of social programs is a key task that policy makers, governments, and program funders perform. In education and the social sciences, a fundamental question asked is ‘How do we know our policy or program works?’

Quantitative Foundations in Educational Research

This course is designed to provide students the fundamental and necessary quantitative methods in educational research.

Data Management (using R)

Good data management is a prerequisite for successful research, needed for reproducibility of results, and essential when collaborating with others.

Tags

Contact