There is an increasing number of datasets with many participants, variables, or both, in education and other fields that often deal with large, multilevel data structures. Once initial confirmatory hypotheses are exhausted, it can be difficult to determine how best to explore the dataset to discover hidden relationships that could help to inform…

Background: This paper considers a new policy and statistical framework for evaluating K-12 schools and policies that aligns with diversity, equity, and inclusion values. The new approach broadens the standard approach to accountability and evaluation by combining features of evaluation and multi-level growth models with approaches used in systems…

Bayesian statistical methods have become more feasible to implement with advances in computing but are not commonly used in educational research. In contrast to frequentist approaches that take hypotheses (and the associated parameters) as fixed, Bayesian methods take data as fixed and hypotheses as random. This difference means that Bayesian…

Modeling the distribution of site level effects is an important problem, but it is also an incredibly difficult one. Current methods rely on distributional assumptions in multilevel models for estimation. There it is hoped that the partial pooling of site level estimates with overall estimates, designed to take into account individual variation as…

The goal of this study is to better understand how methods for estimating treatment effects of latent groups operate. In particular, the authors identify where violations of assumptions can lead to biased estimates, and explore how covariates can be critical in the estimation process. For each set of approaches, the authors first review the…

To date, evaluation research and policy analysis have focused mainly on average program impacts and paid little systematic attention to their variation. Recently, the growing number of multi-site randomized trials that are being planned and conducted make it increasingly feasible to study "cross-site" variation in impacts. Important…

Single case designs (SCDs) are short time series that assess intervention effects by measuring units repeatedly over time both in the presence and absence of treatment. For a variety of reasons, interest in the statistical analysis and meta-analysis of these designs has been growing in recent years. This paper proposes modeling SCD data with…

This study uses simulation examples representing three types of treatment assignment mechanisms in data generation (the random intercept and slopes setting, the random intercept setting, and a third setting with a cluster-level treatment and an individual-level outcome) in order to determine optimal procedures for reducing bias and improving…

This paper presents evidence on the amount of classroom variation in gains across various domains of achievement within the Tennessee Voluntary Pre-Kindergarten (TN-VPK) program; the data are taken from the randomized control trial (RCT) design of the TN-VPK evaluation. This study explores areas where those differences are largest, and offers…

Educators have long used test scores to make educational decisions, but only within the last decade has the availability of data been systematic (Abelman, Elmore, Even, Kenyon, & Marshall, 1999). In recent years, interest has spiked in data-driven decision making in education (Marsh, Pane, & Hamilton, 2006). With technological advances and…

Single-case or single-subject experimental designs (SSED) are used to evaluate the effect of one or more treatments on a single case. Although SSED studies are growing in popularity, the results are in theory case-specific. One systematic and statistical approach for combining single-case data within and across studies is multilevel modeling. The…

Randomized controlled trials (RCTs) are considered the "gold standard" for evaluating an intervention's effectiveness. Recently, the federal government has placed increased emphasis on the use of opportunistic experiments. A key criterion for conducting opportunistic experiments, however, is that there is relatively easy access to data…

Recent guidance on the treatment of missing data in experiments advocates the use of sensitivity analysis and worst-case bounds analysis for addressing non-ignorable missing data mechanisms; moreover, plans for the analysis of missing data should be specified prior to data collection (Puma et al., 2009). While these authors recommend only that…

Access to Head Start has been shown to improve children's preschool experiences and school readiness on selected factors through the end of 1st grade. Two more years of follow-up, through the end of 3rd grade, can now be examined to determine whether these effects continue into the middle elementary grades. The statistical design and impact…

Many community college students begin with the intention of transferring to a four-year school but relatively few actually do. One hypothesis for the low rates of successful two-to-four year transfers is that academic program choices in community colleges are too numerous and too complex. In this paper, the author will address a longer term…